Towards better decoding and language model integration in sequence to sequence models

被引：122

作者：

Chorowski, Jan ^{[1
]}

Jaitly, Navdeep ^{[2
]}

机构：

[1] Google Brain, Mountain View, CA USA

[2] Nvidia, Santa Clara, CA 95051 USA

来源：

18TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2017), VOLS 1-6: SITUATED INTERACTION | 2017年

关键词：

attention mechanism; recurrent neural networks; LSTM;

D O I：

10.21437/Interspeech.2017-343

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

The recently proposed Sequence-to-Sequence (seq2seq) framework advocates replacing complex data processing pipelines, such as an entire automatic speech recognition system, with a single neural network trained in an end-to-end fashion. In this contribution, we analyse an attention-based seq2seq speech recognition system that directly transcribes recordings into characters. We observe two shortcomings: overconfidence in its predictions and a tendency to produce incomplete transcriptions when language models are used. We propose practical solutions to both problems achieving competitive speaker independent word error rates on the Wall Street Journal dataset: without separate language models we reach 10.6% WER, while together with a trigram language model, we reach 6.7% WER. a state-of-the-art result for HMM-free methods.

引用

页码：523 / 527

页数：5