MoonInTheRiver / MoonInTheRiver/DiffSinger
Using higher sample rate for training and inference
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 4.9k
- Forks
- 826
- PR merge metrics
- No merged PRs in 30d
Description
The Diffsinger demo code is great, but the pipeline demonstrated here is all 24k sample rate, is it possible to further improve the sound quality with a higher sampling rate？Does the acoustic model and the HifiGan vocoder have to be retrained if the sample rate is increased to 48k?
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 reviewing the Diffsinger demo pipeline and the sample-rate assumptions in the acoustic model and HiFi-GAN vocoder. Determine whether 48k inference requires retraining either component; done would be a validated answer or documented support for the higher sample rate.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100