babysor / babysor/MockingBird

两种预置声码器各有优缺点,该在什么方向上改进?

Open
#381 1 comment 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
36.9k
Forks
5.2k
PR merge metrics
No merged PRs in 30d

Description

预置的两种声码器g_hifigan和pretained,
用g_hifigan的生成出来的音频,音色特别准,但是会带有电音
用pretained的生成出来的音频,音色没那么准,音量也会变小,但是就不会带电音

这种问题应该往哪个方向去改进?使得结果两种优点都具有

是声码器训练问题?还是源音频的问题?还是合成器?

Contributor guide

No contributing guide indexed for this repository

Research direction

Compare the audio produced by the two preset vocoders, g_hifigan and pretained, and review the issue's possible sources: vocoder training, source audio, or the synthesizer. Done means identifying the cause of the electronic artifacts, reduced timbre accuracy, and lower volume, then defining a direction that preserves the stated advantages of both outputs.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Bug
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.