|
Web
Education
Viewing 1-6 of 6 total results
 Improved Statistical Speech Segmentation Using ...
Improved Statistical Speech Segmentation Using Connectionist Approach 1M.S. Salam, 1Dzulkifli Mohamad and 2S.H. Salleh 1Faculty of Computer Science and Information System, 2Faculty of Biomedical Engineering and Science, University Technology Malaysia, 81310 UTM Skudai, Johor, Malaysia...
SYLLABLE DETECTION AND SEGMENTATION USING TEMPORAL ...
based approach) than a system based purely on phonetic segments [10][15][16]. In view of the above, we believe that segmentation of the acoustic signal into syllabic segments is an important stage in the development of a syllable-centric ASR system. 2. TEMPORAL FLOW MODEL We perform syllabic segmentation using a neural network archi-
(Acoustics Research Department, Bell Laboratories, Murray ...
using voiced-unvoiced-silence contours instead of the measured features. Small but consistent improvements in recognition accuracy have been obtained using these techniques for both speaker-trained and speaker-independent systems over dialed-up telephone lines. 10:20 R8. A statistical approach to metrics for word and syllable recognition.
Isolated Word Recognition System for Malayalam using ...
lated question words from the Malayalam speech query using DWT and ANN and recognition accu-racy of 80% has been reported. 3 Methodology The word identification system designed for Malayalam language uses a syllable based seg-mentation approach. Instead of training the sys-tem with independent words, we use syllable com-...
Prosodic Parameters of French in a Speech Recognition System
However, the way to use prosodie parameters in a speech recognition system is less straightforward than in text-to-speech synthesis. A good predicting model in speech recognition has to forecast different varieties of speaking styles while in speech synthesis one correct speaking style prediction, appropriate to a given application, is sufficient....
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.
|
