Starstruk Logo MarkStarstruk

v2.1.1

PaddlePaddle/PaddleSpeechv2.1.1Aug 16, 2021by zh794390558

AI Summary

A DeepSpeech-focused release that adds alignment utilities, refactors the data pipeline, introduces an online model, and expands dataset support.

Key Highlights

  • Added CTC alignment and MFA alignment examples.
  • Refactored data pipeline and checkpoint save/load mechanisms.
  • Implemented DeepSpeech online model.
  • Added text normalization example and TLG for AISHELL.
  • Expanded datasets to include THCHS30, AIDATATANG, and TIMIT.

Breaking Changes

  • Refactor data pipeline
  • Refactor checkpoint save/load

New Features

  • CTC alignment support
  • MFA alignment example
  • DeepSpeech online model
  • TED en-zh speech translation example
  • Enhanced utils and data processing capabilities

Full Release Notes

1. ctc alignment
2. refactor data pipeline
3. autolog for deepspeech test
4. refactor checkpoint save/load
5. deepspeech online model
6. mfa alignment example
7. add text normaliztion example
8. TLG for aishell
9. more dataest: thchs30, aidatatang, timit etc.
10. 8k speech example
11. ted en-zh st example
12. more utils