Forcing decoder to decode Arabic script only?
Open
- Dominant language
- Python
- Stars
- 40
- Forks
- 3
- PR merge metrics
- No merged PRs in 30d
Description
Hi @zaidalyafeai
Why don't we force the decoder to decode only Arabic script during the evaluation?
[Here](https://github.com/ARBML/whisperar/blob/63769eeaed3c25a2bea2c44fca3c217a8397113f/run_speech_recognition_seq2seq.py#L567)
Contributor guide
No contributing guide indexed for this repository
Research direction
Start at run_speech_recognition_seq2seq.py around line 567 and trace how the decoder is configured during evaluation. Check the surrounding evaluation flow and existing tests or examples, if any. Done requires a clear decision on Arabic-only decoding and an agreed validation path, but the issue does not specify the expected behavior or test coverage.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100