Gated ConvNets for Letter-Based ASR

Vitaliy Liptchinsky, Gabriel Synnaeve, Ronan Collobert

Feb 15, 2018 (modified: Oct 27, 2017) ICLR 2018 Conference Blind Submission readers: everyone Show Bibtex
  • Abstract: In this paper we introduce a new speech recognition system, leveraging a simple letter-based ConvNet acoustic model. The acoustic model requires only audio transcription for training -- no alignment annotations, nor any forced alignment step is needed. At inference, our decoder takes only a word list and a language model, and is fed with letter scores from the acoustic model -- no phonetic word lexicon is needed. Key ingredients for the acoustic model are Gated Linear Units and high dropout. We show near state-of-the-art results in word error rate on the LibriSpeech corpus with MFSC features, both on the clean and other configurations.
  • TL;DR: A letter-based ConvNet acoustic model leads to a simple and competitive speech recognition pipeline.
  • Keywords: automatic speech recognition, letter-based acoustic model, gated convnets
0 Replies