-
funasr的代码封装的太好了,debug fintune代码有点痛苦
有对应的论文吗?直接读论文简单一些
谢谢大佬
-
OS:Linux
Python Version:3.7
Package Version:pytorch==1.13.1、modelscope==1.6.1、funasr==0.6.20
问题描述:
1.用pipeline测试asr:
(1)方法:
inference_pipeline_asr = pipeline(
task=Tasks.auto_speech_rec…
-
Paraformer-large-热词版模型不支持长音频
speech_paraformer-large-contextual_asr_nat-zh-cn-16k-common-vocab8404 识别长音频(7分钟wav文件,大小14M),报错
RuntimeError: [enforce fail at CPUAllocator.cpp:68] . DefaultCPUAllocator:…
-
### 完整错误
```
#0 216.2 Traceback (most recent call last):
#0 216.2 File "/opt/conda/lib/python3.7/site-packages/modelscope/utils/registry.py", line 212, in build_from_cfg
#0 216.2 return ob…
-
os: Centos
Python 3.10.11
Package Version:pytorch 2.0.1、torchaudio 2.0.2、modelscope 1.7.1、funasr 0.5.6 / 0.6.7
Model:'damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch…
-
对比了离线服务和在线服务,以及[sherpa-onnx](https://github.com/k2-fsa/sherpa-onnx/tree/master/sherpa-onnx)的代码,在每个connection连接后,都要根据这个asr_hanlde去创建一个stream,去保证每个线程执行独立,离线转写代码中没这个操作,没太看懂,怎么保证每个线程解码独立的?
https://gith…
-
OS: Linux
Python version: 3.9.16
Torch version: 2.0.1
Torchaudio version: 2.0.2
Modelscope version: 1.7.0
Funasr version: 0.6.7
Model: speech_timestamp_prediction-v1-16k-offline
Command: bash …
-
OS: [e.g. linux]
Python/C++ Version:3.7.16
Package Version:pytorch=1.11.0、torchaudio=0.11.0、modelscope=1.7.1、funasr=0.7.1
Model:speech_paraformer-large-vad-punc_asr_nat-zh-cn-16k-common-vocab8404-p…
-
2024-06-29 16:41:14,003 INFO [export-onnx.py:440] {'best_train_loss': inf, 'best_valid_loss': inf, 'best_train_epoch': -1, 'best_valid_epoch': -1, 'batch_idx_train': 0, 'log_interval': 50, 'reset_inte…
-
OS:ubuntu20.04-cuda11.7.1-py38-torch2.0.1-tf1.15.5-1.7.0Python
Version:3.8.16
Package Version:torchaudio 2.0.2、modelscope 1.7.0、funasr version 0.7.0
Model:speech_paraformer-large_asr_nat-zh-cn-16k…