fish-speech
fish-speech copied to clipboard
Brand new TTS solution
![PixPin_2024-05-05_19-11-10](https://github.com/fishaudio/fish-speech/assets/1288038/aaba86e4-2bdd-4c3e-8702-08bcd940e948)
(fishspeech) [root@a100test fish-speech]# python tools/vqgan/extract_vq.py data --num-workers 1 --batch-size 40 --config-name "vqgan_pretrain" --checkpoint-path "/mnt/fish-speech/checkpoints/vq-gan-group-fsq-2x1024.pth" 2024-05-05 11:15:01.343 | INFO | __main__:main:174 | RANK: 0 / 1 - Starting worker Found 1404...
使用webui推理报错: Traceback (most recent call last): File "/data/workpace/fish-speech/fish_speech/webui/app.py", line 241, in inference resp.raise_for_status() File "/data/workpace/fish-speech/.conda/lib/python3.10/site-packages/requests/models.py", line 1021, in raise_for_status raise HTTPError(http_error_msg, response=self) requests.exceptions.HTTPError: 500 Server Error: for url: http://192.168.1.16:8000/v1/models/default/invoke Traceback...
payload = { "text": text, "reference_text": None, "reference_audio": None, "max_new_tokens": 0, "chunk_length": 30, "top_k": 0, "top_p": 0.7, "repetition_penalty": 1.5, "temperature": 0.7, "speaker": "纳西妲", "format": "wav" } 这样请求还是会有不同说话人的声音,要求:不传入原始音频,仅传入text,固定某一个说话人(女声)
用自有数据预训练+微调 输入文本: 几乎每个人都有颈椎错位,有的人症状轻,有的人症状重,即使自我感觉不明显,越早纠正也越好。 合成音频对应的文本: 几乎每个人都有颈椎错位,有的人症状轻,有的人症状轻,有的人症状轻,有的人症状重,即使自我感觉不明显,越早纠正也越好。
下载 了fishaudio/speech-lm-v1 并手动指定 --tokenizer 为本地文件夹,提示找不到config.json,下载的文件里面也确实没有config.json,在huggingface上也没有找到config.json,请问如何解决呢? raise EnvironmentError( OSError: ./checkpoints does not appear to have a file named config.json. Checkout 'https://huggingface.co/./checkpoints/tree/None' for available files.
Traceback (most recent call last): File "/home/heike/anaconda3/envs/fish-speech/lib/python3.10/site-packages/gradio/queueing.py", line 495, in call_prediction output = await route_utils.call_process_api( File "/home/heike/anaconda3/envs/fish-speech/lib/python3.10/site-packages/gradio/route_utils.py", line 231, in call_process_api output = await app.get_blocks().process_api( File "/home/heike/anaconda3/envs/fish-speech/lib/python3.10/site-packages/gradio/blocks.py", line 1591, in...