Automatic generation of subword units for speech recognition systems

IEEE Transactions on Speech and Audio Processing

Volumn 10, Issue 2, 2002, Pages 89-99

  Singh, Rita ^b   Raj, Bhiksha ^c   Stern, Richard M ^a,b  

^a IEEE

^b CARNEGIE MELLON UNIVERSITY   (United States)

^c MITSUBISHI ELECTRIC RESEARCH LABORATORIES   (United States)

Author keywords

Learning; Lexical representation; Maximum likelihood; Speech recognition; Subword units

Indexed keywords

ACOUSTIC SIGNAL PROCESSING; ALGORITHMS; CLASSIFICATION (OF INFORMATION); GLOSSARIES; MARKOV PROCESSES; MATHEMATICAL MODELS; MAXIMUM LIKELIHOOD ESTIMATION; PROBABILITY DENSITY FUNCTION; SPEECH; SPEECH ANALYSIS; SPEECH SYNTHESIS; TRANSLATION (LANGUAGES);

GAUSSIAN MIXTURES; HIDDEN MARKOV MODELS; MAPPING TABLE; NONIDEOGRAPHIC SCRIPT; OPTIMAL SEGMENTATION; SUBWORD UNITS; TRANSCRIPTIONS;

CONTINUOUS SPEECH RECOGNITION;

EID: 0036476255     PISSN: 10636676     EISSN: None     Source Type: Journal    
DOI: 10.1109/89.985546     Document Type: Article

References (22)

1. An improved subword based speech recognizer
2. Optimizing baseforms for HMM-based speech recognition
3. A method for the construction of acoustic Markov models for words
4. Combined optimization of baseforms and model parameters in speech recognition based on acoustic subword units
5. Joint-lexicon acoustic unit inventory and model design
6. An investigation on the use of acoustic sub-word units for automatic speech recognition
7. Dictionary learning for spontaneous speech recognition
8. Automatic generation of context-dependent pronunciations
9. Automatic generation of German pronunciation variants
10. Incorporation of linguistic knowledge and automatic baseform generation in acoustic subword unit based speech recognition
11. Maximum likelihood from incomplete data via the EM algorithm
12. Learning automata from ordered examples
13. Estimation of probabilities from sparse data for the language model component of a speech recognizer
14. The mathematics of statistical machine translation
15. Automatic clustering and generation of contextual questions for tied states in hidden Markov models
16. The DARPA 1000-Word resource management database for continuous speech recognition
17. Improved acoustic modeling with the SPHINX speech recognition system
18. A statistical text-to-phone function using ngrams and rules