RVC-Project / RVC-Project/Retrieval-based-Voice-Conversion-WebUI
Unable to train feature index file with large datasets.
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 38.4k
- Forks
- 5.3k
- PR merge metrics
- No merged PRs in 30d
Description
I have a 20hr, ~9000 cuts of audio that the model trained fine. At the end of training there was no feature index file created. When trying to train feature index separately the process kills before training begins. Lowering batch size did not help. I was able to train the feature index after reducing the dataset to 5000 cuts.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reproducing separate feature-index training with a dataset near 9,000 audio cuts and compare it with the successful 5,000-cut run. Check why the process terminates before training begins and whether end-to-end training creates the feature index file. Done means large datasets can complete feature-index training or the failure is clearly reported.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100