[Help wanted] Deep Speech 2: End-to-End Speech Recognition in English and Mandarin

[jaeyounkim]

Help wanted: Research paper code and models

Please help us to implement the following research paper using TensorFlow 2.

If you want to contribute, please leave a comment to express your interest.

Research paper

• Deep Speech 2: End-to-End Speech Recognition in English and Mandarin

Status

• In progress

Contributor(s)

• @moneypi

Tasks

• Migrate TensorFlow 1 code to TensorFlow 2
• Convert code to native TensorFlow 2 code

Development environment

• Target
  • Repository: Research directory
  • Directory: https://github.com/tensorflow/models/tree/master/research/deep_speech

Requirements

• Follow our contribution guidelines

License

By contributing, you agree that your contributions will be licensed under Apache License 2.0.

[kyscg]

Reopening issue because native tf2 conversion hasn't been completed @moneypi

[mrinal18]

hello @kyscg and @jaeyounkim , i would like to implement this paper, can you let me what is the process of starting it?

[kyscg]

#8696 converts to temporary tf2 by using compat.v1 workarounds, we will need to migrate the model to native TensorFlow 2.x

The plan right now is:

• Convert AttentionOCR to native TensorFlow 2 and run tests.
• Prepare checkpoints for public usage and consolidate documentation.

More information about conversion to native tf2.x is on the website

[aaryaab]

hey @kyscg I am interested in working on this research paper. Is the issue still open and can I work and contribute for this paper?

[pritamsinha2304]

Is anyone working on it? Is it still open? I can contribute.