The phoneme set influence for lithuanian speech commands recognition accuracy

2017 Open Conference of Electrical, Electronic and Information Sciences (eStream) Pub Date : 2017-04-01 DOI:10.1109/ESTREAM.2017.7950321

M. Greibus, Ž. Ringelienė, L. Telksnys

引用次数: 6

Abstract

The phoneme set influence for Lithuanian speech commands recognition accuracy is investigated. Four phoneme sets are discussed. LIEPA speech corpus for training of Acoustic Model is used. The phonetic representation of corpus transcriptions is generated by grapheme-to-phoneme transformation rules. Rule based transformations for Lithuanian language is proposed. Recognition engine with CMU Pocketsphinx decoder is used. Experimental investigation with fixed size test corpus demonstrates that 36 (FZ1.3) phonemes set shows the bests result.

查看原文本刊更多论文

音位集对立陶宛语语音识别精度的影响

研究了音位集对立陶宛语语音命令识别准确率的影响。讨论了四组音素。使用LIEPA语音语料库进行声学模型训练。语料库转录的语音表示是由字素到音素的转换规则生成的。提出了基于规则的立陶宛语转换。使用带有CMU Pocketsphinx解码器的识别引擎。在固定大小的测试语料库上进行的实验研究表明，36 (FZ1.3)个音素集的效果最好。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

2017 Open Conference of Electrical, Electronic and Information Sciences (eStream)

自引率

0.00%

发文量