OpenNMT / OpenNMT/CTranslate2

Introduce better format for whisper models

Open
#1,741 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
C++
Stars
4.7k
Forks
536
Avg merge
12h 12m
Merged PRs (30d)
4

Description

Currently when using whisper with CTranslate2 we require to provide folder with all the required files such as config.json model.bin etc.
Does it possible to use such file format that will include the decoder, encoder, config files etc all in one file for simplicity?

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reviewing how Whisper models currently require a folder containing config.json, model.bin, and the other required files. Investigate whether the decoder, encoder, and configuration can be represented in one file for CTranslate2. Done means Whisper can use the consolidated format without separately providing the model files.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.