arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1609.03193cs.LGcs.AIcs.CL
收藏

Wav2Letter:一个端到端的ConvNet-based语音识别系统

Wav2Letter: an End-to-End ConvNet-based Speech Recognition System

Facebook AI Research(Facebook AI研究院)

Ronan Collobert, Christian Puhrsch, Gabriel Synnaeve

摘要:

本文提出了一种简单的端到端的语音识别模型,结合基于卷积网络的声学模型和图解码。它被训练输出字母,转录语音,而不需要强制对齐音素。我们引入了一个自动分割标准,用于在没有比对的情况下从序列注释进行训练,该标准与CTC相当,同时更简单。我们在具有MFCC特征的Librispeech语料库上显示出具有竞争力的字出错率结果,并且从原始波形中获得了有希望的结果。

英文摘要:

This paper presents a simple end-to-end model for speech recognition, combining a convolutional network based acoustic model and a graph decoding. It is trained to output letters, with transcribed speech, without the need for force alignment of phonemes. We introduce an automatic segmentation criterion for training from sequence annotation without alignment that is on par with CTC while being simpler. We show competitive results in word error rate on the Librispeech corpus with MFCC features, and promising results from raw waveform.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑