系统仿真学报
繫統倣真學報
계통방진학보
JOURNAL OF SYSTEM SIMULATION
2005年
2期
448-452,456
,共6页
语音转换%模拟电话语音%语音识别%MLLR
語音轉換%模擬電話語音%語音識彆%MLLR
어음전환%모의전화어음%어음식별%MLLR
speech conversion%generated telephone speech%speech recognition%MLLR
提出了一种用语音转换技术改善电话语音识别性能的方法.通过模拟真实电话信道条件下影响语音质量的各种因素,实现由纯净语音到电话语音的转换.识别试验利用模拟电话语音评估了HMM识别器做MLLR自适应前后的性能.实验数据显示,自适应前由转换语音训练的模型识别率比由纯净语音训练的模型识别率增加了18.9%,而自适应试验表明,由转换语音训练而成的模型在MLLR自适应后,系统识别性能进一步得到改善,识别率增加了5.8%.识别实验表明所提语音转换方法可以减小由于真实电话语料缺乏而造成训练语音和测试语音声学性质的不匹配,从而有效地改善电话语音识别系统的性能.
提齣瞭一種用語音轉換技術改善電話語音識彆性能的方法.通過模擬真實電話信道條件下影響語音質量的各種因素,實現由純淨語音到電話語音的轉換.識彆試驗利用模擬電話語音評估瞭HMM識彆器做MLLR自適應前後的性能.實驗數據顯示,自適應前由轉換語音訓練的模型識彆率比由純淨語音訓練的模型識彆率增加瞭18.9%,而自適應試驗錶明,由轉換語音訓練而成的模型在MLLR自適應後,繫統識彆性能進一步得到改善,識彆率增加瞭5.8%.識彆實驗錶明所提語音轉換方法可以減小由于真實電話語料缺乏而造成訓練語音和測試語音聲學性質的不匹配,從而有效地改善電話語音識彆繫統的性能.
제출료일충용어음전환기술개선전화어음식별성능적방법.통과모의진실전화신도조건하영향어음질량적각충인소,실현유순정어음도전화어음적전환.식별시험이용모의전화어음평고료HMM식별기주MLLR자괄응전후적성능.실험수거현시,자괄응전유전환어음훈련적모형식별솔비유순정어음훈련적모형식별솔증가료18.9%,이자괄응시험표명,유전환어음훈련이성적모형재MLLR자괄응후,계통식별성능진일보득도개선,식별솔증가료5.8%.식별실험표명소제어음전환방법가이감소유우진실전화어료결핍이조성훈련어음화측시어음성학성질적불필배,종이유효지개선전화어음식별계통적성능.
A study on speech conversion technology is addressed to improve the telephone speech recognition performance. The speech conversion method is implemented by simulating the influential factors in actual telephone connections. MLLR adaptations are conducted to evaluate the performances of the HMM recognizers, which are trained from the clean speech and generated data respectively. The results without adaptation report that the models trained on generated data can give an 18.9% higher recognition rate than those on clean speech. The adaptation results show that MLLR algorithm contributes an extra increase of 5.8% to the recognition rate of telephone speech system. The experiments illustrate that telephone speech recognition performance can be effectively improved using the generated data, and the conversion method can reduce the acoustic mismatch between the training and test data, which is induced by the shortage of the actual telephone speech.