Advanced search options

Advanced Search Options 🞨

Browse by author name (“Author name starts with…”).

Find ETDs with:

in
/  
in
/  
in
/  
in

Written in Published in Earliest date Latest date

Sorted by

Results per page:

Sorted by: relevance · author · university · dateNew search

University: Hong Kong University of Science and Technology

You searched for subject:(speech processing). Showing records 1 – 30 of 35 total matches.

[1] [2]

Search Limiters

Last 2 Years | English Only

▼ Search Limiters


Hong Kong University of Science and Technology

1. Ko, Yu Ting. Phone deletion modeling in speech recognition.

Degree: 2010, Hong Kong University of Science and Technology

 In a paper published by Greenberg in 1998, it was said that in conversational speech, phone deletion rate may go as high as 12%. On… (more)

Subjects/Keywords: Automatic speech recognition; Phonetics, Acoustic  – Data processing

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Ko, Y. T. (2010). Phone deletion modeling in speech recognition. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b1114779 ; http://repository.ust.hk/ir/bitstream/1783.1-6942/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Ko, Yu Ting. “Phone deletion modeling in speech recognition.” 2010. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b1114779 ; http://repository.ust.hk/ir/bitstream/1783.1-6942/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Ko, Yu Ting. “Phone deletion modeling in speech recognition.” 2010. Web. 15 Oct 2019.

Vancouver:

Ko YT. Phone deletion modeling in speech recognition. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2010. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b1114779 ; http://repository.ust.hk/ir/bitstream/1783.1-6942/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Ko YT. Phone deletion modeling in speech recognition. [Thesis]. Hong Kong University of Science and Technology; 2010. Available from: https://doi.org/10.14711/thesis-b1114779 ; http://repository.ust.hk/ir/bitstream/1783.1-6942/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

2. Ko, Yu Ting. Distinct acoustic modeling for automatic speech recognition.

Degree: 2014, Hong Kong University of Science and Technology

 In triphone-based acoustic modeling, it is difficult to robustly model infrequent triphones due to their lack of training samples. Naive maximum-likelihood (ML) estimation of infrequent… (more)

Subjects/Keywords: Automatic speech recognition; Phonetics, Acoustic; Data processing

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Ko, Y. T. (2014). Distinct acoustic modeling for automatic speech recognition. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b1301519 ; http://repository.ust.hk/ir/bitstream/1783.1-63070/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Ko, Yu Ting. “Distinct acoustic modeling for automatic speech recognition.” 2014. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b1301519 ; http://repository.ust.hk/ir/bitstream/1783.1-63070/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Ko, Yu Ting. “Distinct acoustic modeling for automatic speech recognition.” 2014. Web. 15 Oct 2019.

Vancouver:

Ko YT. Distinct acoustic modeling for automatic speech recognition. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2014. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b1301519 ; http://repository.ust.hk/ir/bitstream/1783.1-63070/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Ko YT. Distinct acoustic modeling for automatic speech recognition. [Thesis]. Hong Kong University of Science and Technology; 2014. Available from: https://doi.org/10.14711/thesis-b1301519 ; http://repository.ust.hk/ir/bitstream/1783.1-63070/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

3. Chan, Daniel Ka-Leung. Automatically merging lexicons that have heterogeneous part-of-speech categories.

Degree: 1999, Hong Kong University of Science and Technology

 Merging lexical resources from different sources is a way to cope with inadequate availability of resources, but lexicon design incompatibilities have been an obstacle to… (more)

Subjects/Keywords: Speech processing systems; Speech synthesis

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Chan, D. K. (1999). Automatically merging lexicons that have heterogeneous part-of-speech categories. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b643173 ; http://repository.ust.hk/ir/bitstream/1783.1-5670/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Chan, Daniel Ka-Leung. “Automatically merging lexicons that have heterogeneous part-of-speech categories.” 1999. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b643173 ; http://repository.ust.hk/ir/bitstream/1783.1-5670/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Chan, Daniel Ka-Leung. “Automatically merging lexicons that have heterogeneous part-of-speech categories.” 1999. Web. 15 Oct 2019.

Vancouver:

Chan DK. Automatically merging lexicons that have heterogeneous part-of-speech categories. [Internet] [Thesis]. Hong Kong University of Science and Technology; 1999. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b643173 ; http://repository.ust.hk/ir/bitstream/1783.1-5670/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Chan DK. Automatically merging lexicons that have heterogeneous part-of-speech categories. [Thesis]. Hong Kong University of Science and Technology; 1999. Available from: https://doi.org/10.14711/thesis-b643173 ; http://repository.ust.hk/ir/bitstream/1783.1-5670/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

4. Li, Chak Fai. Improved polynomial segment model for speech recognition.

Degree: 2004, Hong Kong University of Science and Technology

 While the hidden Markov models (HMM) are the most commonly used models in the field of continuous speech recognition, their weaknesses, such as poor modeling… (more)

Subjects/Keywords: Automatic speech recognition; Speech processing systems

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Li, C. F. (2004). Improved polynomial segment model for speech recognition. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b818955 ; http://repository.ust.hk/ir/bitstream/1783.1-4694/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Li, Chak Fai. “Improved polynomial segment model for speech recognition.” 2004. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b818955 ; http://repository.ust.hk/ir/bitstream/1783.1-4694/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Li, Chak Fai. “Improved polynomial segment model for speech recognition.” 2004. Web. 15 Oct 2019.

Vancouver:

Li CF. Improved polynomial segment model for speech recognition. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2004. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b818955 ; http://repository.ust.hk/ir/bitstream/1783.1-4694/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Li CF. Improved polynomial segment model for speech recognition. [Thesis]. Hong Kong University of Science and Technology; 2004. Available from: https://doi.org/10.14711/thesis-b818955 ; http://repository.ust.hk/ir/bitstream/1783.1-4694/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

5. Chau, Chun Keung. A model-based approach to vocal tract length normalization.

Degree: 2001, Hong Kong University of Science and Technology

 We describe an architecture for speech recognition based interactive toys and discuss a strategy we have adopted to deal with the fact that the speech(more)

Subjects/Keywords: Automatic speech recognition; Speech processing systems

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Chau, C. K. (2001). A model-based approach to vocal tract length normalization. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b712907 ; http://repository.ust.hk/ir/bitstream/1783.1-4599/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Chau, Chun Keung. “A model-based approach to vocal tract length normalization.” 2001. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b712907 ; http://repository.ust.hk/ir/bitstream/1783.1-4599/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Chau, Chun Keung. “A model-based approach to vocal tract length normalization.” 2001. Web. 15 Oct 2019.

Vancouver:

Chau CK. A model-based approach to vocal tract length normalization. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2001. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b712907 ; http://repository.ust.hk/ir/bitstream/1783.1-4599/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Chau CK. A model-based approach to vocal tract length normalization. [Thesis]. Hong Kong University of Science and Technology; 2001. Available from: https://doi.org/10.14711/thesis-b712907 ; http://repository.ust.hk/ir/bitstream/1783.1-4599/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

6. Cheng, Edward Y.C. Context-free grammars and pushdown automata over infinite alphabets.

Degree: 1994, Hong Kong University of Science and Technology

Context-free grammars and pushdown automata over infinite alphabets are defined. They are natural extensions of the classical ones and possess all their properties.

Subjects/Keywords: Speech processing systems; Automatic speech recognition

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Cheng, E. Y. C. (1994). Context-free grammars and pushdown automata over infinite alphabets. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b447462 ; http://repository.ust.hk/ir/bitstream/1783.1-5589/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Cheng, Edward Y C. “Context-free grammars and pushdown automata over infinite alphabets.” 1994. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b447462 ; http://repository.ust.hk/ir/bitstream/1783.1-5589/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Cheng, Edward Y C. “Context-free grammars and pushdown automata over infinite alphabets.” 1994. Web. 15 Oct 2019.

Vancouver:

Cheng EYC. Context-free grammars and pushdown automata over infinite alphabets. [Internet] [Thesis]. Hong Kong University of Science and Technology; 1994. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b447462 ; http://repository.ust.hk/ir/bitstream/1783.1-5589/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Cheng EYC. Context-free grammars and pushdown automata over infinite alphabets. [Thesis]. Hong Kong University of Science and Technology; 1994. Available from: https://doi.org/10.14711/thesis-b447462 ; http://repository.ust.hk/ir/bitstream/1783.1-5589/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

7. Lam, Ying Fung. Cantonese tone recognition using the Hilbert-Huang transform.

Degree: 2014, Hong Kong University of Science and Technology

 Cantonese is a very popular spoken language/dialect, which is well known for its rich set of nine tones and the similarity in tone contours between… (more)

Subjects/Keywords: Automatic speech recognition; Mathematical models; Speech processing systems; Hilbert-Huang transform; Cantonese dialects; Tone

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Lam, Y. F. (2014). Cantonese tone recognition using the Hilbert-Huang transform. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b1274216 ; http://repository.ust.hk/ir/bitstream/1783.1-62794/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Lam, Ying Fung. “Cantonese tone recognition using the Hilbert-Huang transform.” 2014. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b1274216 ; http://repository.ust.hk/ir/bitstream/1783.1-62794/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Lam, Ying Fung. “Cantonese tone recognition using the Hilbert-Huang transform.” 2014. Web. 15 Oct 2019.

Vancouver:

Lam YF. Cantonese tone recognition using the Hilbert-Huang transform. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2014. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b1274216 ; http://repository.ust.hk/ir/bitstream/1783.1-62794/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Lam YF. Cantonese tone recognition using the Hilbert-Huang transform. [Thesis]. Hong Kong University of Science and Technology; 2014. Available from: https://doi.org/10.14711/thesis-b1274216 ; http://repository.ust.hk/ir/bitstream/1783.1-62794/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

8. Tsoi, Pang Kuen. Text-independent speaker identification and verification.

Degree: 2000, Hong Kong University of Science and Technology

 Automatic speaker recognition is one of the major topics in the area of speech recognition. Speaker recognition is the general name for those technologies used… (more)

Subjects/Keywords: Automatic speech recognition; Speech processing systems

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Tsoi, P. K. (2000). Text-independent speaker identification and verification. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b679863 ; http://repository.ust.hk/ir/bitstream/1783.1-4546/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Tsoi, Pang Kuen. “Text-independent speaker identification and verification.” 2000. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b679863 ; http://repository.ust.hk/ir/bitstream/1783.1-4546/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Tsoi, Pang Kuen. “Text-independent speaker identification and verification.” 2000. Web. 15 Oct 2019.

Vancouver:

Tsoi PK. Text-independent speaker identification and verification. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2000. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b679863 ; http://repository.ust.hk/ir/bitstream/1783.1-4546/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Tsoi PK. Text-independent speaker identification and verification. [Thesis]. Hong Kong University of Science and Technology; 2000. Available from: https://doi.org/10.14711/thesis-b679863 ; http://repository.ust.hk/ir/bitstream/1783.1-4546/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

9. Lam, Wai Bun. Optimizing the performance of GPD-based discriminative training in speech recognition.

Degree: 2000, Hong Kong University of Science and Technology

Speech Recognition is becoming more important in our daily life. Many applications are starting to use this technology to make them more effective. One important… (more)

Subjects/Keywords: Automatic speech recognition; Speech processing systems

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Lam, W. B. (2000). Optimizing the performance of GPD-based discriminative training in speech recognition. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b679875 ; http://repository.ust.hk/ir/bitstream/1783.1-4558/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Lam, Wai Bun. “Optimizing the performance of GPD-based discriminative training in speech recognition.” 2000. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b679875 ; http://repository.ust.hk/ir/bitstream/1783.1-4558/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Lam, Wai Bun. “Optimizing the performance of GPD-based discriminative training in speech recognition.” 2000. Web. 15 Oct 2019.

Vancouver:

Lam WB. Optimizing the performance of GPD-based discriminative training in speech recognition. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2000. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b679875 ; http://repository.ust.hk/ir/bitstream/1783.1-4558/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Lam WB. Optimizing the performance of GPD-based discriminative training in speech recognition. [Thesis]. Hong Kong University of Science and Technology; 2000. Available from: https://doi.org/10.14711/thesis-b679875 ; http://repository.ust.hk/ir/bitstream/1783.1-4558/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

10. Ma, Chi Yuen. Multilingual speech recognition and its application in a multilingual voice browser.

Degree: 2001, Hong Kong University of Science and Technology

 Hong Kong is a two literature three language (Cantonese, Mandarin and English) city. The people speak in a mixed-code manner-many English terms are used in… (more)

Subjects/Keywords: Automatic speech recognition; Speech processing systems; Speech synthesis

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Ma, C. Y. (2001). Multilingual speech recognition and its application in a multilingual voice browser. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b712897 ; http://repository.ust.hk/ir/bitstream/1783.1-4589/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Ma, Chi Yuen. “Multilingual speech recognition and its application in a multilingual voice browser.” 2001. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b712897 ; http://repository.ust.hk/ir/bitstream/1783.1-4589/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Ma, Chi Yuen. “Multilingual speech recognition and its application in a multilingual voice browser.” 2001. Web. 15 Oct 2019.

Vancouver:

Ma CY. Multilingual speech recognition and its application in a multilingual voice browser. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2001. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b712897 ; http://repository.ust.hk/ir/bitstream/1783.1-4589/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Ma CY. Multilingual speech recognition and its application in a multilingual voice browser. [Thesis]. Hong Kong University of Science and Technology; 2001. Available from: https://doi.org/10.14711/thesis-b712897 ; http://repository.ust.hk/ir/bitstream/1783.1-4589/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

11. Cattle, Andrew Grant CSE. Improving humour recognition using mental lexicons, world knowledge, and joke structure.

Degree: 2018, Hong Kong University of Science and Technology

 As natural language interfaces become more prevalent, the ability for computers to both understand and create humour becomes more important. Humour is a ubiquitous part… (more)

Subjects/Keywords: Natural language processing (Computer science); Speech perception; Computational linguistics; Wit and humor; Data processing

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Cattle, A. G. C. (2018). Improving humour recognition using mental lexicons, world knowledge, and joke structure. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-991012671057703412 ; http://repository.ust.hk/ir/bitstream/1783.1-97733/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Cattle, Andrew Grant CSE. “Improving humour recognition using mental lexicons, world knowledge, and joke structure.” 2018. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-991012671057703412 ; http://repository.ust.hk/ir/bitstream/1783.1-97733/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Cattle, Andrew Grant CSE. “Improving humour recognition using mental lexicons, world knowledge, and joke structure.” 2018. Web. 15 Oct 2019.

Vancouver:

Cattle AGC. Improving humour recognition using mental lexicons, world knowledge, and joke structure. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2018. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-991012671057703412 ; http://repository.ust.hk/ir/bitstream/1783.1-97733/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Cattle AGC. Improving humour recognition using mental lexicons, world knowledge, and joke structure. [Thesis]. Hong Kong University of Science and Technology; 2018. Available from: https://doi.org/10.14711/thesis-991012671057703412 ; http://repository.ust.hk/ir/bitstream/1783.1-97733/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

12. Dey, Anik. Acoustic modeling for Hindi speech recognition under low resource conditions.

Degree: 2014, Hong Kong University of Science and Technology

 We show for the first time how to use a multilingual acoustic modeling approach for Hindi speech recognition under low resource conditions using deep neural… (more)

Subjects/Keywords: Automatic speech recognition; Phonetics, Acoustic; Data processing; Hindi language

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Dey, A. (2014). Acoustic modeling for Hindi speech recognition under low resource conditions. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b1347268 ; http://repository.ust.hk/ir/bitstream/1783.1-71876/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Dey, Anik. “Acoustic modeling for Hindi speech recognition under low resource conditions.” 2014. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b1347268 ; http://repository.ust.hk/ir/bitstream/1783.1-71876/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Dey, Anik. “Acoustic modeling for Hindi speech recognition under low resource conditions.” 2014. Web. 15 Oct 2019.

Vancouver:

Dey A. Acoustic modeling for Hindi speech recognition under low resource conditions. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2014. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b1347268 ; http://repository.ust.hk/ir/bitstream/1783.1-71876/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Dey A. Acoustic modeling for Hindi speech recognition under low resource conditions. [Thesis]. Hong Kong University of Science and Technology; 2014. Available from: https://doi.org/10.14711/thesis-b1347268 ; http://repository.ust.hk/ir/bitstream/1783.1-71876/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

13. Tam, Yik-cheung. Development of an asynchronous multi-band system for continuous speech recognition.

Degree: 2001, Hong Kong University of Science and Technology

 Recently, multi-band automatic speech recognition (MBASR) has been proposed by Bourlard et al. and Hermansky et al. to improve robustness under noisy environment. It is… (more)

Subjects/Keywords: Automatic speech recognition; Signal processing

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Tam, Y. (2001). Development of an asynchronous multi-band system for continuous speech recognition. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b706547 ; http://repository.ust.hk/ir/bitstream/1783.1-5714/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Tam, Yik-cheung. “Development of an asynchronous multi-band system for continuous speech recognition.” 2001. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b706547 ; http://repository.ust.hk/ir/bitstream/1783.1-5714/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Tam, Yik-cheung. “Development of an asynchronous multi-band system for continuous speech recognition.” 2001. Web. 15 Oct 2019.

Vancouver:

Tam Y. Development of an asynchronous multi-band system for continuous speech recognition. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2001. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b706547 ; http://repository.ust.hk/ir/bitstream/1783.1-5714/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Tam Y. Development of an asynchronous multi-band system for continuous speech recognition. [Thesis]. Hong Kong University of Science and Technology; 2001. Available from: https://doi.org/10.14711/thesis-b706547 ; http://repository.ust.hk/ir/bitstream/1783.1-5714/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

14. Chan, Arthur Yu Chung. Robust speech recognition against unknown short-time noise.

Degree: 2002, Hong Kong University of Science and Technology

 The performance of automatic speech recognition (ASR) is known to degrade under noise corruption. Such means of corruption include additive noises such as air-conditioning noise,… (more)

Subjects/Keywords: Automatic speech recognition; Signal processing

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Chan, A. Y. C. (2002). Robust speech recognition against unknown short-time noise. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b775112 ; http://repository.ust.hk/ir/bitstream/1783.1-4637/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Chan, Arthur Yu Chung. “Robust speech recognition against unknown short-time noise.” 2002. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b775112 ; http://repository.ust.hk/ir/bitstream/1783.1-4637/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Chan, Arthur Yu Chung. “Robust speech recognition against unknown short-time noise.” 2002. Web. 15 Oct 2019.

Vancouver:

Chan AYC. Robust speech recognition against unknown short-time noise. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2002. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b775112 ; http://repository.ust.hk/ir/bitstream/1783.1-4637/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Chan AYC. Robust speech recognition against unknown short-time noise. [Thesis]. Hong Kong University of Science and Technology; 2002. Available from: https://doi.org/10.14711/thesis-b775112 ; http://repository.ust.hk/ir/bitstream/1783.1-4637/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

15. Zhang, Jian. Extractive speech summarization using structural modeling.

Degree: 2011, Hong Kong University of Science and Technology

 In this dissertation, we propose a new rhetorical structure modeling approach as a critical step in the understanding of extractive summarization of spoken documents. Previous… (more)

Subjects/Keywords: Automatic speech recognition; Natural language processing (Computer science)

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Zhang, J. (2011). Extractive speech summarization using structural modeling. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b1155523 ; http://repository.ust.hk/ir/bitstream/1783.1-7330/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Zhang, Jian. “Extractive speech summarization using structural modeling.” 2011. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b1155523 ; http://repository.ust.hk/ir/bitstream/1783.1-7330/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Zhang, Jian. “Extractive speech summarization using structural modeling.” 2011. Web. 15 Oct 2019.

Vancouver:

Zhang J. Extractive speech summarization using structural modeling. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2011. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b1155523 ; http://repository.ust.hk/ir/bitstream/1783.1-7330/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Zhang J. Extractive speech summarization using structural modeling. [Thesis]. Hong Kong University of Science and Technology; 2011. Available from: https://doi.org/10.14711/thesis-b1155523 ; http://repository.ust.hk/ir/bitstream/1783.1-7330/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

16. Wong, Shuen Kong. Channel and noise adaptation via HMM mixture mean transform and stochastic matching.

Degree: 1998, Hong Kong University of Science and Technology

 Automatic speech recognition is very sensitive to mismatch between training and testing condition, eg, differences in background noise level and data acquisition equipment. The recognition… (more)

Subjects/Keywords: Automatic speech recognition; Signal processing

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Wong, S. K. (1998). Channel and noise adaptation via HMM mixture mean transform and stochastic matching. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b603897 ; http://repository.ust.hk/ir/bitstream/1783.1-4484/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Wong, Shuen Kong. “Channel and noise adaptation via HMM mixture mean transform and stochastic matching.” 1998. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b603897 ; http://repository.ust.hk/ir/bitstream/1783.1-4484/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Wong, Shuen Kong. “Channel and noise adaptation via HMM mixture mean transform and stochastic matching.” 1998. Web. 15 Oct 2019.

Vancouver:

Wong SK. Channel and noise adaptation via HMM mixture mean transform and stochastic matching. [Internet] [Thesis]. Hong Kong University of Science and Technology; 1998. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b603897 ; http://repository.ust.hk/ir/bitstream/1783.1-4484/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Wong SK. Channel and noise adaptation via HMM mixture mean transform and stochastic matching. [Thesis]. Hong Kong University of Science and Technology; 1998. Available from: https://doi.org/10.14711/thesis-b603897 ; http://repository.ust.hk/ir/bitstream/1783.1-4484/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

17. Zhu, Wenya CSE. Generative dialogue system with knowledge base.

Degree: 2018, Hong Kong University of Science and Technology

 Building dialogue systems that can converse naturally with humans is a challenging yet intriguing problem of artificial intelligence. Dialogue system is categorized into task-oriented system… (more)

Subjects/Keywords: Machine learning; Data processing; Speech; Interactive computer systems; Human-computer interaction

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Zhu, W. C. (2018). Generative dialogue system with knowledge base. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-991012636669103412 ; http://repository.ust.hk/ir/bitstream/1783.1-95786/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Zhu, Wenya CSE. “Generative dialogue system with knowledge base.” 2018. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-991012636669103412 ; http://repository.ust.hk/ir/bitstream/1783.1-95786/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Zhu, Wenya CSE. “Generative dialogue system with knowledge base.” 2018. Web. 15 Oct 2019.

Vancouver:

Zhu WC. Generative dialogue system with knowledge base. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2018. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-991012636669103412 ; http://repository.ust.hk/ir/bitstream/1783.1-95786/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Zhu WC. Generative dialogue system with knowledge base. [Thesis]. Hong Kong University of Science and Technology; 2018. Available from: https://doi.org/10.14711/thesis-991012636669103412 ; http://repository.ust.hk/ir/bitstream/1783.1-95786/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

18. Wang, Linlin ECE. Disfluency detection and reconstruction of spontaneous speech transcripts.

Degree: 2016, Hong Kong University of Science and Technology

 Disfluencies in spoken language remain a challenge to both language processing applications and human perception. Disfluency identification and removal are therefore beneficial steps to improve… (more)

Subjects/Keywords: Automatic speech recognition; Natural language processing (Computer science)

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Wang, L. E. (2016). Disfluency detection and reconstruction of spontaneous speech transcripts. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b1626114 ; http://repository.ust.hk/ir/bitstream/1783.1-93204/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Wang, Linlin ECE. “Disfluency detection and reconstruction of spontaneous speech transcripts.” 2016. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b1626114 ; http://repository.ust.hk/ir/bitstream/1783.1-93204/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Wang, Linlin ECE. “Disfluency detection and reconstruction of spontaneous speech transcripts.” 2016. Web. 15 Oct 2019.

Vancouver:

Wang LE. Disfluency detection and reconstruction of spontaneous speech transcripts. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2016. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b1626114 ; http://repository.ust.hk/ir/bitstream/1783.1-93204/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Wang LE. Disfluency detection and reconstruction of spontaneous speech transcripts. [Thesis]. Hong Kong University of Science and Technology; 2016. Available from: https://doi.org/10.14711/thesis-b1626114 ; http://repository.ust.hk/ir/bitstream/1783.1-93204/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

19. Wong, Hon Wah. Variable speed playback system for speech and audio signals (and topics in video processing).

Degree: 1998, Hong Kong University of Science and Technology

 For many audio/speech retrieval applications, it is desirable to perform time scale modification (TSM) to change the apparent rhythm of audio and human speech. Such… (more)

Subjects/Keywords: Speech processing systems; Signal processing; Image processing

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Wong, H. W. (1998). Variable speed playback system for speech and audio signals (and topics in video processing). (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b603890 ; http://repository.ust.hk/ir/bitstream/1783.1-4478/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Wong, Hon Wah. “Variable speed playback system for speech and audio signals (and topics in video processing).” 1998. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b603890 ; http://repository.ust.hk/ir/bitstream/1783.1-4478/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Wong, Hon Wah. “Variable speed playback system for speech and audio signals (and topics in video processing).” 1998. Web. 15 Oct 2019.

Vancouver:

Wong HW. Variable speed playback system for speech and audio signals (and topics in video processing). [Internet] [Thesis]. Hong Kong University of Science and Technology; 1998. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b603890 ; http://repository.ust.hk/ir/bitstream/1783.1-4478/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Wong HW. Variable speed playback system for speech and audio signals (and topics in video processing). [Thesis]. Hong Kong University of Science and Technology; 1998. Available from: https://doi.org/10.14711/thesis-b603890 ; http://repository.ust.hk/ir/bitstream/1783.1-4478/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

20. Liu, Wai Kat. Fast accent classification and adaptation.

Degree: 2000, Hong Kong University of Science and Technology

 Automatic speech recognition technology has developed rapidly in the past decade. Applications of this technology have been seen everywhere such as voice dialing in mobile… (more)

Subjects/Keywords: Automatic speech recognition; Speech processing systems; Accents and accentuation

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Liu, W. K. (2000). Fast accent classification and adaptation. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b673094 ; http://repository.ust.hk/ir/bitstream/1783.1-4543/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Liu, Wai Kat. “Fast accent classification and adaptation.” 2000. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b673094 ; http://repository.ust.hk/ir/bitstream/1783.1-4543/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Liu, Wai Kat. “Fast accent classification and adaptation.” 2000. Web. 15 Oct 2019.

Vancouver:

Liu WK. Fast accent classification and adaptation. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2000. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b673094 ; http://repository.ust.hk/ir/bitstream/1783.1-4543/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Liu WK. Fast accent classification and adaptation. [Thesis]. Hong Kong University of Science and Technology; 2000. Available from: https://doi.org/10.14711/thesis-b673094 ; http://repository.ust.hk/ir/bitstream/1783.1-4543/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

21. Au, Wing Hei. Improved acoustic model training for speech recognition and verification.

Degree: 2004, Hong Kong University of Science and Technology

 Model estimation is an important step in pattern recognition tasks. Different optimization criteria, such as maximum likelihood (ML), minimum classification error (MCE) and maximum mutual… (more)

Subjects/Keywords: Automatic speech recognition; Speech processing systems  – Mathematical models

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Au, W. H. (2004). Improved acoustic model training for speech recognition and verification. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b842122 ; http://repository.ust.hk/ir/bitstream/1783.1-4718/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Au, Wing Hei. “Improved acoustic model training for speech recognition and verification.” 2004. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b842122 ; http://repository.ust.hk/ir/bitstream/1783.1-4718/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Au, Wing Hei. “Improved acoustic model training for speech recognition and verification.” 2004. Web. 15 Oct 2019.

Vancouver:

Au WH. Improved acoustic model training for speech recognition and verification. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2004. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b842122 ; http://repository.ust.hk/ir/bitstream/1783.1-4718/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Au WH. Improved acoustic model training for speech recognition and verification. [Thesis]. Hong Kong University of Science and Technology; 2004. Available from: https://doi.org/10.14711/thesis-b842122 ; http://repository.ust.hk/ir/bitstream/1783.1-4718/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

22. Lai, Yiu Pong. Maximum likelihood normalization for robust speech recognition.

Degree: 2003, Hong Kong University of Science and Technology

 One important issue in speech recognition is the ability to handle variations caused by unseen speakers or noise. One approach to handle variations is to… (more)

Subjects/Keywords: Automatic speech recognition; Speech processing systems  – Mathematical models

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Lai, Y. P. (2003). Maximum likelihood normalization for robust speech recognition. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b807392 ; http://repository.ust.hk/ir/bitstream/1783.1-4666/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Lai, Yiu Pong. “Maximum likelihood normalization for robust speech recognition.” 2003. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b807392 ; http://repository.ust.hk/ir/bitstream/1783.1-4666/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Lai, Yiu Pong. “Maximum likelihood normalization for robust speech recognition.” 2003. Web. 15 Oct 2019.

Vancouver:

Lai YP. Maximum likelihood normalization for robust speech recognition. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2003. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b807392 ; http://repository.ust.hk/ir/bitstream/1783.1-4666/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Lai YP. Maximum likelihood normalization for robust speech recognition. [Thesis]. Hong Kong University of Science and Technology; 2003. Available from: https://doi.org/10.14711/thesis-b807392 ; http://repository.ust.hk/ir/bitstream/1783.1-4666/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

23. Chong, Fong Ho. Frequency-stream-tying hidden Markov model.

Degree: 2003, Hong Kong University of Science and Technology

 In most speech recognition systems, acoustic features are extracted from the whole frequency spectrum of a speech signal. Corruption of the speech signal in one… (more)

Subjects/Keywords: Speech processing systems; Automatic speech recognition; Markov processes

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Chong, F. H. (2003). Frequency-stream-tying hidden Markov model. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b786960 ; http://repository.ust.hk/ir/bitstream/1783.1-4657/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Chong, Fong Ho. “Frequency-stream-tying hidden Markov model.” 2003. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b786960 ; http://repository.ust.hk/ir/bitstream/1783.1-4657/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Chong, Fong Ho. “Frequency-stream-tying hidden Markov model.” 2003. Web. 15 Oct 2019.

Vancouver:

Chong FH. Frequency-stream-tying hidden Markov model. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2003. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b786960 ; http://repository.ust.hk/ir/bitstream/1783.1-4657/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Chong FH. Frequency-stream-tying hidden Markov model. [Thesis]. Hong Kong University of Science and Technology; 2003. Available from: https://doi.org/10.14711/thesis-b786960 ; http://repository.ust.hk/ir/bitstream/1783.1-4657/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

24. Chan, Siu Man. Improved speaker verification with discrimination power weighting.

Degree: 2004, Hong Kong University of Science and Technology

 Biometric authentication has become increasingly popular. Speaker verification, or sometimes called voice print, authenticates speaker identity by voice. One major advantage of speaker verification is… (more)

Subjects/Keywords: Automatic speech recognition; Speech processing systems; Biometric identification

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Chan, S. M. (2004). Improved speaker verification with discrimination power weighting. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b842087 ; http://repository.ust.hk/ir/bitstream/1783.1-4701/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Chan, Siu Man. “Improved speaker verification with discrimination power weighting.” 2004. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b842087 ; http://repository.ust.hk/ir/bitstream/1783.1-4701/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Chan, Siu Man. “Improved speaker verification with discrimination power weighting.” 2004. Web. 15 Oct 2019.

Vancouver:

Chan SM. Improved speaker verification with discrimination power weighting. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2004. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b842087 ; http://repository.ust.hk/ir/bitstream/1783.1-4701/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Chan SM. Improved speaker verification with discrimination power weighting. [Thesis]. Hong Kong University of Science and Technology; 2004. Available from: https://doi.org/10.14711/thesis-b842087 ; http://repository.ust.hk/ir/bitstream/1783.1-4701/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

25. Shun, Louis Lok Yee. Speaker verification with discriminative vector quantization.

Degree: 1998, Hong Kong University of Science and Technology

 Vector Quantization (VQ) has been used widely in the areas of data compression and digital signal processing. Speaker identification and verification systems based on VQ… (more)

Subjects/Keywords: Vector analysts; Automatic speech recognition; Speech processing systems

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Shun, L. L. Y. (1998). Speaker verification with discriminative vector quantization. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b583786 ; http://repository.ust.hk/ir/bitstream/1783.1-5654/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Shun, Louis Lok Yee. “Speaker verification with discriminative vector quantization.” 1998. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b583786 ; http://repository.ust.hk/ir/bitstream/1783.1-5654/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Shun, Louis Lok Yee. “Speaker verification with discriminative vector quantization.” 1998. Web. 15 Oct 2019.

Vancouver:

Shun LLY. Speaker verification with discriminative vector quantization. [Internet] [Thesis]. Hong Kong University of Science and Technology; 1998. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b583786 ; http://repository.ust.hk/ir/bitstream/1783.1-5654/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Shun LLY. Speaker verification with discriminative vector quantization. [Thesis]. Hong Kong University of Science and Technology; 1998. Available from: https://doi.org/10.14711/thesis-b583786 ; http://repository.ust.hk/ir/bitstream/1783.1-5654/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

26. Liu, Yi. Pronunciation modeling for spontaneous mandarin speech recognition.

Degree: 2002, Hong Kong University of Science and Technology

 The focus in automatic speech recognition (ASR) research has gradually shifted from read speech to spontaneous speech. ASR systems can reach an accuracy of above… (more)

Subjects/Keywords: Automatic speech recognition; Speech processing systems; Chinese language  – Data processing

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Liu, Y. (2002). Pronunciation modeling for spontaneous mandarin speech recognition. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b775795 ; http://repository.ust.hk/ir/bitstream/1783.1-570/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Liu, Yi. “Pronunciation modeling for spontaneous mandarin speech recognition.” 2002. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b775795 ; http://repository.ust.hk/ir/bitstream/1783.1-570/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Liu, Yi. “Pronunciation modeling for spontaneous mandarin speech recognition.” 2002. Web. 15 Oct 2019.

Vancouver:

Liu Y. Pronunciation modeling for spontaneous mandarin speech recognition. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2002. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b775795 ; http://repository.ust.hk/ir/bitstream/1783.1-570/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Liu Y. Pronunciation modeling for spontaneous mandarin speech recognition. [Thesis]. Hong Kong University of Science and Technology; 2002. Available from: https://doi.org/10.14711/thesis-b775795 ; http://repository.ust.hk/ir/bitstream/1783.1-570/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

27. Ng, Kwong Tim. Exploring Chinese linguistic characteristics for speech recognition.

Degree: 2005, Hong Kong University of Science and Technology

 During the past decades, automatic speech recognition technology has been applied in different commercial sectors around the world, including China. However, Chinese speech recognition systems… (more)

Subjects/Keywords: Automatic speech recognition; Speech processing systems; Chinese language  – Data processing

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Ng, K. T. (2005). Exploring Chinese linguistic characteristics for speech recognition. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b917841 ; http://repository.ust.hk/ir/bitstream/1783.1-4745/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Ng, Kwong Tim. “Exploring Chinese linguistic characteristics for speech recognition.” 2005. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b917841 ; http://repository.ust.hk/ir/bitstream/1783.1-4745/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Ng, Kwong Tim. “Exploring Chinese linguistic characteristics for speech recognition.” 2005. Web. 15 Oct 2019.

Vancouver:

Ng KT. Exploring Chinese linguistic characteristics for speech recognition. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2005. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b917841 ; http://repository.ust.hk/ir/bitstream/1783.1-4745/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Ng KT. Exploring Chinese linguistic characteristics for speech recognition. [Thesis]. Hong Kong University of Science and Technology; 2005. Available from: https://doi.org/10.14711/thesis-b917841 ; http://repository.ust.hk/ir/bitstream/1783.1-4745/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

28. Yim, Chi Ho. Auditory spectrum based features (ASBF) for robust speech recognition.

Degree: 2001, Hong Kong University of Science and Technology

Speech recognition is the enabling technology allowing humans to communicate with computers using their voices. While many speech recognizers work well in laboratory environment, their… (more)

Subjects/Keywords: Automatic speech recognition; Computers sound processing; Signal processing

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Yim, C. H. (2001). Auditory spectrum based features (ASBF) for robust speech recognition. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b696753 ; http://repository.ust.hk/ir/bitstream/1783.1-4583/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Yim, Chi Ho. “Auditory spectrum based features (ASBF) for robust speech recognition.” 2001. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b696753 ; http://repository.ust.hk/ir/bitstream/1783.1-4583/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Yim, Chi Ho. “Auditory spectrum based features (ASBF) for robust speech recognition.” 2001. Web. 15 Oct 2019.

Vancouver:

Yim CH. Auditory spectrum based features (ASBF) for robust speech recognition. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2001. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b696753 ; http://repository.ust.hk/ir/bitstream/1783.1-4583/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Yim CH. Auditory spectrum based features (ASBF) for robust speech recognition. [Thesis]. Hong Kong University of Science and Technology; 2001. Available from: https://doi.org/10.14711/thesis-b696753 ; http://repository.ust.hk/ir/bitstream/1783.1-4583/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

29. Ye, Guoli. The use of discrete distributions with a very large codebook for automatic speech recognition and speaker verification.

Degree: 2013, Hong Kong University of Science and Technology

 With the advance of semiconductor technology and the popularity of distributed speech/speaker recognition paradigm (e.g., Siri in iPhone4s), here we revisit the use of discrete… (more)

Subjects/Keywords: Automatic speech recognition; Markov processes; Information storage and retrieval systems; Data processing

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Ye, G. (2013). The use of discrete distributions with a very large codebook for automatic speech recognition and speaker verification. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b1213370 ; http://repository.ust.hk/ir/bitstream/1783.1-7887/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Ye, Guoli. “The use of discrete distributions with a very large codebook for automatic speech recognition and speaker verification.” 2013. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b1213370 ; http://repository.ust.hk/ir/bitstream/1783.1-7887/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Ye, Guoli. “The use of discrete distributions with a very large codebook for automatic speech recognition and speaker verification.” 2013. Web. 15 Oct 2019.

Vancouver:

Ye G. The use of discrete distributions with a very large codebook for automatic speech recognition and speaker verification. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2013. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b1213370 ; http://repository.ust.hk/ir/bitstream/1783.1-7887/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Ye G. The use of discrete distributions with a very large codebook for automatic speech recognition and speaker verification. [Thesis]. Hong Kong University of Science and Technology; 2013. Available from: https://doi.org/10.14711/thesis-b1213370 ; http://repository.ust.hk/ir/bitstream/1783.1-7887/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation


Hong Kong University of Science and Technology

30. Lo, Yuen-Yee. Colloquial language modeling for spontaneous Cantonese speech recognition.

Degree: 2000, Hong Kong University of Science and Technology

 This thesis presents a colloquial language modeling technique for spontaneous Cantonese speech recognition. Cantonese is a language linguistically distinct from standard written Chinese. It is… (more)

Subjects/Keywords: Speech processing systems; Cantonese dialects  – Data processing; Cantonese dialects  – China  – Hong Kong  – Slang

Record DetailsSimilar RecordsGoogle PlusoneFacebookTwitterCiteULikeMendeleyreddit

APA · Chicago · MLA · Vancouver · CSE | Export to Zotero / EndNote / Reference Manager

APA (6th Edition):

Lo, Y. (2000). Colloquial language modeling for spontaneous Cantonese speech recognition. (Thesis). Hong Kong University of Science and Technology. Retrieved from https://doi.org/10.14711/thesis-b655900 ; http://repository.ust.hk/ir/bitstream/1783.1-4539/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Chicago Manual of Style (16th Edition):

Lo, Yuen-Yee. “Colloquial language modeling for spontaneous Cantonese speech recognition.” 2000. Thesis, Hong Kong University of Science and Technology. Accessed October 15, 2019. https://doi.org/10.14711/thesis-b655900 ; http://repository.ust.hk/ir/bitstream/1783.1-4539/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

MLA Handbook (7th Edition):

Lo, Yuen-Yee. “Colloquial language modeling for spontaneous Cantonese speech recognition.” 2000. Web. 15 Oct 2019.

Vancouver:

Lo Y. Colloquial language modeling for spontaneous Cantonese speech recognition. [Internet] [Thesis]. Hong Kong University of Science and Technology; 2000. [cited 2019 Oct 15]. Available from: https://doi.org/10.14711/thesis-b655900 ; http://repository.ust.hk/ir/bitstream/1783.1-4539/1/th_redirect.html.

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

Council of Science Editors:

Lo Y. Colloquial language modeling for spontaneous Cantonese speech recognition. [Thesis]. Hong Kong University of Science and Technology; 2000. Available from: https://doi.org/10.14711/thesis-b655900 ; http://repository.ust.hk/ir/bitstream/1783.1-4539/1/th_redirect.html

Note: this citation may be lacking information needed for this citation format:
Not specified: Masters Thesis or Doctoral Dissertation

[1] [2]

.