Pitch dependent phone modelling for HMM-based speech recognition.
スポンサーリンク
概要
- 論文の詳細を見る
This paper proposes a novel method of incorporating pitch information into an HMM speech recognition system by exploiting the correlation between pitch and spectral parameters, <I>e.g</I>. cepstrum. Pitch patterns are not used explicitly; instead, spectral parameters are normalized framewise according to the pitch value. Evidence is given to show that the use of pitch information consistently improves the recognition performance. Experiments with 24 phoneme labels showed that the phoneme error rate for fast continuous speech could be improved by about 10%. Using these pitchnormalized phone models in an HMM-LR speech recognition system improved the phrase recognition accuracy for the top 5 candidates from 96% to 97.5%, <I>i.e</I>. the error rate was nearly halved.
- 一般社団法人 日本音響学会の論文
一般社団法人 日本音響学会 | 論文
- How large is the individual difference in hearing sensitivity?: Establishment of ISO 28961 on the statistical distribution of hearing thresholds of otologically normal young persons
- Applying generation process model constraint to fundamental frequency contours generated by hidden-Markov-model-based speech synthesis
- Vocal cord vibration in the production of consonants. Observation by means of high-speed digital imaging using a fiberscope.:Observation by means of high-speed digital imaging using a fiberscope
- The early reflections of the impulse response in an auditorium.
- Multiple reflections between rigid plane panels.