zipformer

Table of contents

• 1. Description
• 2. Current Support Platform
• 3. Pretrained Model
• 4. Convert to RKNN
• 5. Python Demo
• 6. Android Demo
  • 6.1 Compile and Build
  • 6.2 Push demo files to device
  • 6.3 Run demo
• 7. Linux Demo
  • 7.1 Compile && Build
  • 7.2 Push demo files to device
  • 7.3 Run demo
• 8. Expected Results

1. Description

Chinese-English ASR model using k2-zipformer-streaming.

The model used in this example comes from the following open source projects:

https://huggingface.co/csukuangfj/k2fsa-zipformer-bilingual-zh-en-t

2. Current Support Platform

RK3562, RK3566, RK3568, RK3576, RK3588, RV1126B

3. Pretrained Model

Download link:

encoder-epoch-99-avg-1.onnx
decoder-epoch-99-avg-1.onnx
joiner-epoch-99-avg-1.onnx

Download with shell command:

cd model
./download_model.sh

Note: For exporting zipformer onnx models, please refer to export-for-onnx.sh

4. Convert to RKNN

Usage:

cd python
python convert.py <onnx_model> <TARGET_PLATFORM> <dtype(optional)> <output_rknn_path(optional)>

# such as: 
python convert.py ../model/encoder-epoch-99-avg-1.onnx rk3588
# output model will be saved as ../model/encoder-epoch-99-avg-1.rknn

python convert.py ../model/decoder-epoch-99-avg-1.onnx rk3588
# output model will be saved as ../model/decoder-epoch-99-avg-1.rknn

python convert.py ../model/joiner-epoch-99-avg-1.onnx rk3588
# output model will be saved as ../model/joiner-epoch-99-avg-1.rknn

Description:

• <onnx_model>: Specify ONNX model path.
• <TARGET_PLATFORM>: Specify NPU platform name. Support Platform refer here.
• <dtype>(optional): Specify as i8 or fp. i8 for doing quantization, fp for no quantization. Default is fp.
• <output_rknn_path>(optional): Specify save path for the RKNN model, default save in the same directory as ONNX model.

5. Python Demo

Usage:

# Install kaldifeat
# Refer to https://csukuangfj.github.io/kaldifeat/installation/from_wheels.html for installation.
# This python demo is tested under version: kaldifeat-1.25.4.dev20240223

cd python
# Inference with ONNX model
python zipformer.py --encoder_model_path <onnx_model> --decoder_model_path <onnx_model> ---joiner_model_path <onnx_model>

# Inference with RKNN model
python zipformer.py --encoder_model_path <rknn_model> --decoder_model_path <rknn_model> --joiner_model_path <rknn_model> --target <TARGET_PLATFORM>

Description:

• <TARGET_PLATFORM>: Specify NPU platform name. Support Platform refer here.
• <onnx_model / rknn_model>: Specify model path.

6. Android Demo

6.1 Compile and Build

Usage:

# go back to the rknn_model_zoo root directory
cd ../../
export ANDROID_NDK_PATH=<android_ndk_path>

./build-android.sh -t <TARGET_PLATFORM> -a <ARCH> -d zipformer

# such as 
./build-android.sh -t rk3588 -a arm64-v8a -d zipformer

Description:

• <android_ndk_path>: Specify Android NDK path.
• <TARGET_PLATFORM>: Specify NPU platform name. Support Platform refer here.
• <ARCH>: Specify device system architecture. To query device architecture, refer to the following command:

  # Query architecture. For Android, ['arm64-v8a' or 'armeabi-v7a'] should shown in log.
  adb shell cat /proc/version
  

6.2 Push demo files to device

With device connected via USB port, push demo files to devices:

adb root
adb remount
adb push install/<TARGET_PLATFORM>_android_<ARCH>/rknn_zipformer_demo/ /data/

6.3 Run demo

adb shell
cd /data/rknn_zipformer_demo

export LD_LIBRARY_PATH=./lib
./rknn_zipformer_demo model/encoder-epoch-99-avg-1.rknn model/decoder-epoch-99-avg-1.rknn model/joiner-epoch-99-avg-1.rknn model/test.wav

7. Linux Demo

Please note that the Linux compilation tool chain recommends using gcc-linaro-6.3.1(aarch64)/gcc-arm-8.3(armhf)/armhf-uclibcgnueabihf(armhf for RV1106/RV1103 series). Using other versions may encounter the problem of Cdemo compilation failure. For detailed compilation guide, please refer to Compilation_Environment_Setup_Guide.md

7.1 Compile and Build

usage

# go back to the rknn_model_zoo root directory
cd ../../

# if GCC_COMPILER not found while building, please set GCC_COMPILER path
(optional)export GCC_COMPILER=<GCC_COMPILER_PATH>

./build-linux.sh -t <TARGET_PLATFORM> -a <ARCH> -d zipformer

# such as 
./build-linux.sh -t rk3588 -a aarch64 -d zipformer

Description:

• <GCC_COMPILER_PATH>: Specified as GCC_COMPILER path.

• <TARGET_PLATFORM> : Specify NPU platform name. Support Platform refer here.

• <ARCH>: Specify device system architecture. To query device architecture, refer to the following command:

  # Query architecture. For Linux, ['aarch64' or 'armhf'] should shown in log.
  adb shell cat /proc/version
  

7.2 Push demo files to device

• If device connected via USB port, push demo files to devices:

adb push install/<TARGET_PLATFORM>_linux_<ARCH>/rknn_zipformer_demo/ /data/

• For other boards, use scp or other approaches to push all files under install/<TARGET_PLATFORM>_linux_<ARCH>/rknn_zipformer_demo/ to data.

7.3 Run demo

adb shell
cd /data/rknn_zipformer_demo

export LD_LIBRARY_PATH=./lib
./rknn_zipformer_demo model/encoder-epoch-99-avg-1.rknn model/decoder-epoch-99-avg-1.rknn model/joiner-epoch-99-avg-1.rknn model/test.wav

8. Expected Results

This example will print the recognized text, as follows:

Zipformer output: 对我做了介绍那么我想说的是大家如果对我的研究感兴趣呢

• Note: Different platforms, different versions of tools and drivers may have slightly different results.