Using prosodic and lexical information for speaker identification

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

Volumn 1, Issue , 2002, Pages 141-144

  Weber, Frederick ^a   Manganaro, Linda ^a   Peskin, Barbara ^a,b   Shriberg, Elizabeth ^b,c  

^a Dragon Systems/Lernout and Hauspie   (United States)

^b INTERNATIONAL COMPUTER SCIENCE INSTITUTE   (United States)

^c SRI INTERNATIONAL   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

GAUSSIAN MIXTURE MODELS; LARGE VOCABULARY CONTINUOUS SPEECH RECOGNITION; LEXICAL INFORMATION; PROSODIC INFORMATION;

ACOUSTIC SIGNAL PROCESSING; DATABASE SYSTEMS; LARGE SCALE SYSTEMS; MATHEMATICAL MODELS; SPEECH ANALYSIS; SPEECH COMMUNICATION; TELEPHONE SYSTEMS;

SPEECH RECOGNITION;

EID: 0036299157     PISSN: 15206149     EISSN: None     Source Type: Journal    
DOI: 10.1109/ICASSP.2002.5743674     Document Type: Article

References (12)

1. Robust text-independent speaker identification using Gaussian mixture speaker models
2. Speaker recognition on single- and multi-speaker data
3. Switchboard: Telephone speech corpus for research and development
4. Prosody-based automatic segmentation of speech into sentences and topics
5. Modeling dynamic prosodic variation for speaker verification
6. Automatic speaker recognition based on pitch contours
7. Speaker recognition based on idiolectal differences between speakers
8. Comparison of background normalization methods for text-independent speaker verification
9. Improvements in recognition of conversational telephone speech
10. The DET curve in assessment of detection task performance
11. Disfluency annotation stylebook for the switchboard corpus