Several methods were developed to improve grapheme-to-phoneme （G2P） conversion models for Chinese text-to-speech （TTS） systems. The critical problem of data sparsity was handled by combining approaches. First, a t...Several methods were developed to improve grapheme-to-phoneme （G2P） conversion models for Chinese text-to-speech （TTS） systems. The critical problem of data sparsity was handled by combining approaches. First, a text-selection method was designed to cover as many G2P text corpus contexts as possible. Then, various data-driven modeling methods were used with comparisons to select the best method for each polyphonic word. Finally, independent models were used for some neutral tone words in addition to the normal G2P models to achieve more compact and flexible G2P models. Tests show that these methods reduce the relative errors by 50% for both normal polyphonic words and Chinese neutral tones.展开更多
针对蒙古文字母到音素的转换(grapheme to phoneme conversion,G2P)问题,提出了基于规则的蒙古文G2P转换方法和基于联合序列模型的蒙古文G2P转换方法。实验结果表明,利用联合序列模型的蒙古文G2P转换方法要明显好于基于规则的蒙古文G2P...针对蒙古文字母到音素的转换(grapheme to phoneme conversion,G2P)问题,提出了基于规则的蒙古文G2P转换方法和基于联合序列模型的蒙古文G2P转换方法。实验结果表明,利用联合序列模型的蒙古文G2P转换方法要明显好于基于规则的蒙古文G2P转换方法。并且建立的基于联合序列模型的蒙古文G2P转换系统的词误识率为16.32%,音素误识率仅为3.37%,能达到实用要求。展开更多
在字母语言的文本到语音(text to speech,TTS)系统和自动语音识别(automatic speech recognition,ASR)系统的实际应用中,由于新词的出现和大量专有名词收集、存储的困难,经常会遇到发音词典中不存在的单词,需要自动对这些单词注音。该...在字母语言的文本到语音(text to speech,TTS)系统和自动语音识别(automatic speech recognition,ASR)系统的实际应用中,由于新词的出现和大量专有名词收集、存储的困难,经常会遇到发音词典中不存在的单词,需要自动对这些单词注音。该文对各类自动注音方法做了回顾,对自动注音系统中若干关键问题做了简单的介绍。在此基础上,该文给出了一个自动注音系统的实现,对其中若干技术细节对性能的影响做了分析,最后给出了本系统与基于手写规则的方法的对比实验,结果表明前者的单词转换正确率达63%以上,显著优于后者。展开更多
文摘Several methods were developed to improve grapheme-to-phoneme （G2P） conversion models for Chinese text-to-speech （TTS） systems. The critical problem of data sparsity was handled by combining approaches. First, a text-selection method was designed to cover as many G2P text corpus contexts as possible. Then, various data-driven modeling methods were used with comparisons to select the best method for each polyphonic word. Finally, independent models were used for some neutral tone words in addition to the normal G2P models to achieve more compact and flexible G2P models. Tests show that these methods reduce the relative errors by 50% for both normal polyphonic words and Chinese neutral tones.
文摘针对蒙古文字母到音素的转换(grapheme to phoneme conversion,G2P)问题,提出了基于规则的蒙古文G2P转换方法和基于联合序列模型的蒙古文G2P转换方法。实验结果表明,利用联合序列模型的蒙古文G2P转换方法要明显好于基于规则的蒙古文G2P转换方法。并且建立的基于联合序列模型的蒙古文G2P转换系统的词误识率为16.32%,音素误识率仅为3.37%,能达到实用要求。
文摘在字母语言的文本到语音(text to speech,TTS)系统和自动语音识别(automatic speech recognition,ASR)系统的实际应用中,由于新词的出现和大量专有名词收集、存储的困难,经常会遇到发音词典中不存在的单词,需要自动对这些单词注音。该文对各类自动注音方法做了回顾,对自动注音系统中若干关键问题做了简单的介绍。在此基础上,该文给出了一个自动注音系统的实现,对其中若干技术细节对性能的影响做了分析,最后给出了本系统与基于手写规则的方法的对比实验,结果表明前者的单词转换正确率达63%以上,显著优于后者。
