Facebook 开源语音识别工具包 wav2letter

Linuxeden 开源社区 --

Facebook AI 研究院近日开源了一款简单高效的端到端自动语音识别（ASR）系统 wav2letter，wav2letter 实现的是论文 Wav2Letter: an End-to-End ConvNet-based Speech Recognition System 和及 Letter-Based Speech Recognition with Gated ConvNets 中提出的架构。

Papers

@article{collobert:2016,
  author    = {Ronan Collobert and Christian Puhrsch and Gabriel Synnaeve},
  title     = {Wav2Letter: an End-to-End ConvNet-based Speech Recognition System},
  journal   = {CoRR},
  volume    = {abs/1609.03193},
  year      = {2016},
  url       = {http://ift.tt/2cSDMfp},
}

和

@article{liptchinsky:2017,
  author    = {Vitaliy Liptchinsky and Gabriel Synnaeve and Ronan Collobert},
  title     = {Letter-Based Speech Recognition with Gated ConvNets},
  journal   = {CoRR},
  volume    = {abs/1712.09444},
  year      = {2017},
  url       = {http://ift.tt/2ChyusY},
}

如果你使用 wav2letter 或相关的预训练模型，需引用其中的一篇论文。

另外，如果想要立刻进行语音转录的，Facebook 还提供了 Librispeech 数据集上 预训练 模型。

转自 http://ift.tt/2Etlj6B

The post Facebook 开源语音识别工具包 wav2letter appeared first on Linuxeden开源社区.

http://ift.tt/2lBC1YK