RVC-Project/Retrieval-based-Voice-Conversion-WebUI

Generated audio became blurred.

Open

#292 opened on May 16, 2023

View on GitHub
 (1 comment) (0 reactions) (0 assignees)Python (2,849 forks)batch import
bughelp wanted

Repository metrics

Stars
 (18,427 stars)
PR merge metrics
 (No merged PRs in 30d)

Description

I am training with COLAB, but the voice of the model trained yesterday and today seems noisier and blurred compared to the voice of the previously trained model. It feels like the pre-trained models are not working. The former is the model learned previously and the latter is the model learned today. https://www.axfc.net/u/4065483 Although trained on the same data set, the sound is clearly degraded in the latter. https://www.axfc.net/u/4065492 I tried learning at 40k and it was clear. I tried learning at 48k and it seemed to cause problems.

Contributor guide

Research direction
Investigate the training pipeline for 48kHz sample rate. Compare the pre trained models used for 40kHz and 48kHz. Check for any upsampling/downsampling issues or differences in preprocessing.
Tech stack
python
Domain
machine learning
Issue type
Bug
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-2 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Clear
Prerequisites
PythonBasic machine learning knowledge
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
40