Skip to content

Repository files navigation

Code release for the paper: Accurate, fast, cheap, choose three!

We borrow or rely on:

  1. a codebase derived from WeNet e2e repo
  2. derived code from Mamba-SSM
  3. derived code from optimized_transducer
  4. The fstalign tool

License

Instructions

Look at examples/gigaspeech/s0/README_RevPaper_Choose3.md for details

About

Repository for code behind the Interspeech 2025 paper : Accurate, fast, cheap: Choose three. Replacing Multi-Head-Attention with Bidirectional Recurrent Attention for Long-Form ASR

Resources

Code of conduct

Contributing

Stars

10 stars

Watchers

1 watching

Forks

Releases

Packages

Used by

Contributors

Languages