|
Web
Education
US6076057A - Unsupervised HMM adaptation based on speech ...
An unsupervised, discriminative, sentence level, HMM adaptation based on speech-silence classification is presented. Silence and speech regions are determined either using a speech end-pointer or the segmentation obtained from the recognizer in a first pass. The discriminative training procedure using a GPD or any other discriminative training algorithm, employed in conjunction with the HMM ......
IEEE/ACM TRANSACTIONS ON AUDIO, SPEECH, AND LANGUAGE ...
the separation is [26]. Hence, the use of separate dictionaries for each speech unit provides a more precise representation in the high-dimensional feature space. Training frame sequences representing various speech units (speech exemplars) are extracted based on the state-level align-ments obtained using a conventional HMM-based recognizer.
Symmetry | Free Full-Text | Evaluation of Speech Quality ...
This paper discusses an approach for assessing the quality of speech while undergoing speech rehabilitation. One of the main reasons for speech quality decrease during the surgical treatment of vocal tract diseases is the loss of the vocal tractˈs parts and the disruption of its symmetry. In particular, one of the most common oncological diseases of the oral cavity is cancer of the tongue....
Laboratory for Intelligent Multimedia Processing (IMP)
Speaker diarization is the process of segmentation of speech files to speakers, tracking of a certain speaker in a given file and speaker tying (tracking of a given speaker in different files which may be recorded in different environments). Speech segmentation using modified Bayes Information Criterion
Speech Recognition by Machine: A Review - arXiv.org e ...
speech recognition systems, it is necessary to build statistical models for sub word speech units, build up word models from these sub word speech unit models (using a lexicon to describe the composition of words), and then postulate word sequences and evaluate the acoustic model probabilities via standard concatenation methods.
Recent Research Towards Advanced Man-Machine Interface ...
Furthermore, a sequential speaker adaptation, using input speech itself, attained a recognition rate of 97.4% for the male speakers. Finally, a speaker normalization scheme based on fuzzy mapping with the adapted codebook was found to be effective for a non-SPLIT based recognition system.
 Bibliography - UPSpace
[44] R.J. Di Francesco, "Real-time speech segmentation using pitch and convexity jump models: Application to variable rate speech coding," IEEE Transactions on Acoustics, Speech, and Signal Processing, vol. 38, no. 5, pp. 741-748, May
Introduction - York University
· Statistical Part-of-Speech Tagging ... - Input via DTMF or high-accuracy connected digit recognition ... pairs are grouped to train routing matrix using a data-driven approach · Technologies: speech recognition, language modeling, call routing, dialogue generation .
|
