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

Why does my RVC model produces bad results?

Open

#2,025 opened on 2024年4月29日

GitHub で見る
 (1 comment) (0 reactions) (0 assignees)Python (18,427 stars) (2,849 forks)batch import
help wanted

説明

I created a RVC model and I had it set to 50 epochs for training. It produces bad results. Here is the audio file I used for training, I used version v2 and I set the target sample rate to 48k. https://files.catbox.moe/440g5r.wav Here is the input file I used. https://vocaroo.com/11ANq5W9xFd5 Here is the output file. https://vocaroo.com/136PewYoZmL3

Edit: I discovered Seed-VC which worked well with 6 minutes of training data. It was also worked with 14 seconds of training data.

コントリビューターガイド

技術スタック
pythonpytorch
領域
machine learning
Issue 種別
research
難度新しい貢献者にとっての実装難度です。1 は非常に小さな変更、5 は専門的な作業です。
4
推定時間経験ある貢献者が調査、実装、テスト、pull request 準備にかけるおおよその時間範囲です。
under 1 hour
活動状況Issue が今どれだけ取り組みやすいかを示します。新しい、活発、古い、ブロック中、またはメンテナー入力待ちなどです。
fresh
明確さ期待される変更、受け入れ条件、次の手順がどれだけ明確かを示します。
unclear
前提条件
PythonPyTorchRVC basicsVoice conversion fundamentals
初心者向け度初回貢献者にどれだけ取り組みやすいかを 1-100 で推定したスコアです。
20
調査方針
The issue reports poor RVC training results. To investigate, start by checking the provided training audio (440g5r.wav) for quality and length, examine the training hyperparameters (50 epochs, target sample rate 48k), and compare with the working Seed VC approach. The repo's training scripts (e.g., train.py) and configuration files should be reviewed to identify potential misconfigurations. Since no maintainer has replied, a comment asking for the specific training command and logs would clarify whether this is a user error or a software bug.