1.5B模型在1min左右开始有几率发出啸叫声
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 37.8k
- Forks
- 4.3k
- Avg merge
- 7m
- Merged PRs (30d)
- 1
Description
如题，第一次测试的时候默认设置，cfg2，步数10，字数2000左右，直出了6min无克隆的音频特别完美，当时很兴奋
可惜的是，之后的测试再也没有复现这个壮举，基本上在1min左右开始出现微弱啸叫声，3-4min左右变大无法接受
目前我的测试下，cfg1.5，是比较稳定的1min左右可以稳定没有啸叫，或者啸叫可以接受。步数的增加似乎没有能够有效控制啸叫，流失输出与非流失输出差别也不大
不过我除了cfg和步数外，并没有修改其他参数，是否有其他参数的调整可以增强稳定性？
其实对比0.5B的版本，已经是巨大的进步了，当时30s左右就开始啸叫，现在稳定在1min左右，特别棒！！
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
No source file or test is named. Start by reproducing the 1.5B model's long-generation behavior with the reported defaults, then compare cfg1.5 and cfg2, step counts, and dropout settings. Done means identifying a reproducible stability cause or documenting a parameter combination that prevents unacceptable howling.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- audio-video-rtc, machine-learning
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 48/100