Open LuWu9 opened 1 month ago
如何在torch 2.4.0环境下使用sensevoice 我正在做一个纯本地运行的语音对话的项目,使用SenseVoice+LLM+GPT-SoVITS来达到语音对语音交流的效果,其中部分模块必须要使用torch2.4.0,所以求大佬帮忙看下SenseVoice这个项目怎么样才能在torch2.4.0环境下正常运行 异常见下图: ①运行在纯净的虚拟环境下,仅将torch升级为2.4.0,python版本3.11.2; ②torchaudio的版本与torch同步; ③尝试过修改weights_only=True,除了少了几行提示其他没有区别)
weights_only=True
from funasr import AutoModel from funasr.utils.postprocess_utils import rich_transcription_postprocess import time class ASR: def __init__(self): # 本地模型路径 model_dir = "./models/SenseVoiceSmall" vad_model_dir = "./models/speech_fsmn_vad_zh-cn-16k-common-pytorch" self.model = AutoModel( model=model_dir, trust_remote_code=True, remote_code="./src/SenseVoice/model.py", vad_model=vad_model_dir, vad_kwargs={"max_single_segment_time": 30000}, device="cuda:0", ) def audio2text(self, filepath): res = self.model.generate( input=filepath, cache={}, language="zh", use_itn=True, batch_size_s=60, merge_vad=True, merge_length_s=15, ) text = rich_transcription_postprocess(res[0]["text"]) return text if __name__ == "__main__": filepath = "./.cache/Keira.wav" a = ASR() start_time = time.time() text = a.audio2text(filepath) print("{0}({1:.2f}s)".format(text, time.time()-start_time))
使用虚拟环境pip install -r requirements.txt后可以正常输出(除torch和torchaudio版本外,环境与上方异常情况相同)
Package Version ---------------------- ------------ aiofiles 23.2.1 aliyun-python-sdk-core 2.15.1 aliyun-python-sdk-kms 2.16.3 annotated-types 0.7.0 antlr4-python3-runtime 4.9.3 anyio 4.4.0 audioread 3.0.1 certifi 2024.7.4 cffi 1.16.0 charset-normalizer 3.3.2 click 8.1.7 colorama 0.4.6 contourpy 1.2.1 crcmod 1.7 cryptography 43.0.0 cycler 0.12.1 decorator 5.1.1 editdistance 0.8.1 fastapi 0.112.0 ffmpy 0.4.0 filelock 3.15.4 fonttools 4.53.1 fsspec 2024.6.1 funasr 1.1.4 gradio 4.40.0 gradio_client 1.2.0 h11 0.14.0 httpcore 1.0.5 httpx 0.27.0 huggingface 0.0.1 huggingface-hub 0.24.5 hydra-core 1.3.2 idna 3.7 importlib_resources 6.4.0 intel-openmp 2021.4.0 jaconv 0.4.0 jamo 0.4.1 jieba 0.42.1 Jinja2 3.1.4 jmespath 0.10.0 joblib 1.4.2 kaldiio 2.18.0 kiwisolver 1.4.5 lazy_loader 0.4 librosa 0.10.2.post1 llvmlite 0.43.0 markdown-it-py 3.0.0 MarkupSafe 2.1.5 matplotlib 3.9.1 mdurl 0.1.2 mkl 2021.4.0 modelscope 1.17.0 mpmath 1.3.0 msgpack 1.0.8 networkx 3.3 numba 0.60.0 numpy 1.26.4 omegaconf 2.3.0 orjson 3.10.6 oss2 2.18.6 packaging 24.1 pandas 2.2.2 pillow 10.4.0 pip 22.3.1 platformdirs 4.2.2 pooch 1.8.2 protobuf 5.27.3 pycparser 2.22 pycryptodome 3.20.0 pydantic 2.8.2 pydantic_core 2.20.1 pydub 0.25.1 Pygments 2.18.0 pynndescent 0.5.13 pyparsing 3.1.2 python-dateutil 2.9.0.post0 python-multipart 0.0.9 pytorch-wpe 0.0.1 pytz 2024.1 PyYAML 6.0.1 requests 2.32.3 rich 13.7.1 ruff 0.5.6 scikit-learn 1.5.1 scipy 1.14.0 semantic-version 2.10.0 sentencepiece 0.2.0 setuptools 65.5.0 shellingham 1.5.4 six 1.16.0 sniffio 1.3.1 soundfile 0.12.1 soxr 0.4.0 starlette 0.37.2 sympy 1.13.1 tbb 2021.13.0 tensorboardX 2.6.2.2 threadpoolctl 3.5.0 tomlkit 0.12.0 torch 2.4.0 torch-complex 0.4.4 torchaudio 2.4.0 tqdm 4.66.5 typer 0.12.3 typing_extensions 4.12.2 tzdata 2024.1 umap-learn 0.5.6 urllib3 2.2.2 uvicorn 0.30.5 websockets 12.0
❓ Questions and Help
What is your question?
如何在torch 2.4.0环境下使用sensevoice 我正在做一个纯本地运行的语音对话的项目,使用SenseVoice+LLM+GPT-SoVITS来达到语音对语音交流的效果,其中部分模块必须要使用torch2.4.0,所以求大佬帮忙看下SenseVoice这个项目怎么样才能在torch2.4.0环境下正常运行 异常见下图: ①运行在纯净的虚拟环境下,仅将torch升级为2.4.0,python版本3.11.2; ②torchaudio的版本与torch同步; ③尝试过修改
weights_only=True
,除了少了几行提示其他没有区别)Code
What have you tried?
使用虚拟环境pip install -r requirements.txt后可以正常输出(除torch和torchaudio版本外,环境与上方异常情况相同)
What's your environment?