voxcom infer 偶现不稳定,有时候会推理完4096token(超过2分钟),
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 37.8k
- Forks
- 4.3k
- Avg merge
- 7m
- Merged PRs (30d)
- 1
Description
如图的位置,voxcpm.py VoxCPMModel inference 有时候会推理完4096token(超过2分钟),如果把max_length设置的比较小比如1024,推理15s后结果音频是乱的且长度也不对应,在我的prompt_wav和prompt_text不对应时,会经常出现(概率大概有5%),当他们对应时,也会偶尔出现几次(不过概率比较低< 1%)
我的配置:voxcpm 1.0.5 torch==2.6.0 H20卡
我的部分demo:
self.tts_model = VoxCPM(voxcpm_model_path=self.model_dir) def generate(self, text, prompt_wav_path, prompt_text=None, cfg=2.0, steps=10): try: wav = self.tts_model.generate( text=text, prompt_wav_path=prompt_wav_path, prompt_text=prompt_text, cfg_value=cfg, inference_timesteps=steps, normalize=False, denoise=False, retry_badcase=False, max_length=1024, ) return 16000, wav
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 with voxcpm.py and the VoxCPMModel inference path, then reproduce the supplied generate call with max_length=1024 and mismatched prompt_wav and prompt_text. Compare the occasional 4096-token runs with the malformed, incorrectly sized audio results; done means the intermittent behavior is reproduced and its cause and correction are documented.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, pytorch
- Domain
- audio-video-rtc, machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100