speech_recognition/FunASR
1
0
Fork 0
mirror of https://github.com/modelscope/FunASR synced 2025-09-15 14:48:36 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

docs

Browse Source
This commit is contained in:
游雁 2023-04-21 15:32:34 +08:00
parent 5cfb12c7da
commit 14a14d0e40
8 changed files with 50 additions and 32 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

0
docs/recipe/asr_recipe.md → docs/academic_recipe/asr_recipe.md
View File

0
docs/recipe/lm_recipe.md → docs/academic_recipe/lm_recipe.md
View File

0
docs/recipe/punc_recipe.md → docs/academic_recipe/punc_recipe.md
View File

0
docs/recipe/sd_recipe.md → docs/academic_recipe/sd_recipe.md
View File

0
docs/recipe/sv_recipe.md → docs/academic_recipe/sv_recipe.md
View File

0
docs/recipe/vad_recipe.md → docs/academic_recipe/vad_recipe.md
View File

18
docs/index.rst
View File

@ -22,13 +22,13 @@ Overview
.. toctree::
:maxdepth: 1
:caption: Academic Recipe
:caption: Academic Egs
./recipe/asr_recipe.md
./recipe/punc_recipe.md
./recipe/vad_recipe.md
./recipe/sv_recipe.md
./recipe/sd_recipe.md
./academic_recipe/asr_recipe.md
./academic_recipe/punc_recipe.md
./academic_recipe/vad_recipe.md
./academic_recipe/sv_recipe.md
./academic_recipe/sd_recipe.md
.. toctree::
@ -40,7 +40,7 @@ Overview
.. toctree::
:maxdepth: 1
:caption: ModelScope pipeline
:caption: ModelScope Egs
./modescope_pipeline/quick_start.md
./modescope_pipeline/asr_pipeline.md
@ -52,13 +52,13 @@ Overview
.. toctree::
:maxdepth: 1
:caption: Huggingface pipeline
:caption: Huggingface Egs
Undo
.. toctree::
:maxdepth: 1
:caption: funasr library
:caption: Funasr Library
./build_task.md

64
funasr/export/README.md
View File

@ -16,7 +16,7 @@ pip install torch-quant # Optional, for torchscript quantization
pip install onnx onnxruntime # Optional, for onnx quantization
```
## Export model
## Usage
`Tips`: torch>=1.11.0
```shell
@ -37,7 +37,46 @@ pip install onnx onnxruntime # Optional, for onnx quantization
`fallback-num`: specify the number of fallback layers to perform automatic mixed precision quantization.
## Performance Benchmark of Runtime
### Export onnx format model
#### Export model from modelscope
```shell
python -m funasr.export.export_model --model-name damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch --export-dir ./export --type onnx
```
#### Export model from local path
The model'name must be `model.pb`
```shell
python -m funasr.export.export_model --model-name /mnt/workspace/damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch --export-dir ./export --type onnx
```
#### Test onnx model
Ref to [test](https://github.com/alibaba-damo-academy/FunASR/tree/main/funasr/export/test)
### Export torchscripts format model
#### Export model from modelscope
```shell
python -m funasr.export.export_model --model-name damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch --export-dir ./export --type torch
```
#### Export model from local path
The model'name must be `model.pb`
```shell
python -m funasr.export.export_model --model-name /mnt/workspace/damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch --export-dir ./export --type torch
```
#### Test onnx model
Ref to [test](https://github.com/alibaba-damo-academy/FunASR/tree/main/funasr/export/test)
## Runtime
### ONNXRuntime
#### ONNXRuntime-python
Ref to [docs](https://alibaba-damo-academy.github.io/FunASR/en/runtime/onnxruntime_python.html)
#### ONNXRuntime-cpp
Ref to [docs](https://alibaba-damo-academy.github.io/FunASR/en/runtime/onnxruntime_cpp.html)
### Libtorch
#### Libtorch-python
Ref to [docs](https://alibaba-damo-academy.github.io/FunASR/en/runtime/libtorch_python.html)
#### Libtorch-cpp
Undo
## Performance Benchmark
### Paraformer on CPU
@ -48,27 +87,6 @@ pip install onnx onnxruntime # Optional, for onnx quantization
### Paraformer on GPU
[nv-triton](https://github.com/alibaba-damo-academy/FunASR/tree/main/funasr/runtime/triton_gpu)
## For example
### Export onnx format model
Export model from modelscope
```shell
python -m funasr.export.export_model --model-name damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch --export-dir ./export --type onnx
```
Export model from local path, the model'name must be `model.pb`.
```shell
python -m funasr.export.export_model --model-name /mnt/workspace/damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch --export-dir ./export --type onnx
```
### Export torchscripts format model
Export model from modelscope
```shell
python -m funasr.export.export_model --model-name damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch --export-dir ./export --type torch
```
Export model from local path, the model'name must be `model.pb`.
```shell
python -m funasr.export.export_model --model-name /mnt/workspace/damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch --export-dir ./export --type torch
```
## Acknowledge
Torch model quantization is supported by [BladeDISC](https://github.com/alibaba/BladeDISC), an end-to-end DynamIc Shape Compiler project for machine learning workloads. BladeDISC provides general, transparent, and ease of use performance optimization for TensorFlow/PyTorch workloads on GPGPU and CPU backends. If you are interested, please contact us.