产品上手指引
Linux PC
AI 加速卡
LLM-8850 Card
大语言模型
实时 AI 语音助手
火山引擎语音助手
离线语音识别
工业控制
Ezdata
Ethernet 摄像头
PoECAM
Wi-Fi 摄像头
Unit CamS3/-5MP
AI 摄像头
LoRa & LoRaWAN
电机驱动
恢复出厂固件教程
SenseVoice
SenseVoice 是一种语音识别与理解模型，能够高效、准确地将语音内容转化为文本，并支持多语言和多场景的语音处理。
- 手动下载模型 并上传到 raspberrypi5，或者通过以下命令拉取模型仓库。
提示
如果没有安装 git lfs，先参考git lfs 安装说明进行安装。
git clone https://huggingface.co/AXERA-TECH/SenseVoice文件说明：
m5stack@raspberrypi:~/rsp/SenseVoice $ ls -lh
total 464K
-rw-rw-r-- 1 m5stack m5stack 11K Aug 12 16:38 am.mvn
-rw-rw-r-- 1 m5stack m5stack 369K Aug 12 16:38 chn_jpn_yue_eng_ko_spectok.bpe.model
-rw-rw-r-- 1 m5stack m5stack 0 Aug 12 16:38 config.json
-rw-rw-r-- 1 m5stack m5stack 108 Aug 12 16:38 download_dataset.sh
-rw-rw-r-- 1 m5stack m5stack 893 Aug 12 16:38 download_utils.py
drwxrwxr-x 2 m5stack m5stack 4.0K Aug 12 16:38 embeddings
-rw-rw-r-- 1 m5stack m5stack 17K Aug 12 16:38 frontend.py
-rw-rw-r-- 1 m5stack m5stack 1.1K Aug 12 16:38 LICENSE
-rw-rw-r-- 1 m5stack m5stack 1.6K Aug 12 16:38 main.py
-rw-rw-r-- 1 m5stack m5stack 3.2K Aug 12 16:38 print_utils.py
-rw-rw-r-- 1 m5stack m5stack 1.5K Aug 12 16:38 README.md
-rw-rw-r-- 1 m5stack m5stack 71 Aug 12 16:38 requirements.txt
drwxrwxr-x 2 m5stack m5stack 4.0K Aug 12 16:38 sensevoice_ax650
-rw-rw-r-- 1 m5stack m5stack 9.1K Aug 12 16:38 SenseVoiceAx.py
-rw-rw-r-- 1 m5stack m5stack 2.5K Aug 12 16:38 test_wer.py
-rw-rw-r-- 1 m5stack m5stack 4.7K Aug 12 16:38 tokenizer.py- 创建虚拟环境
python -m venv sensevoice- 激活虚拟环境
source sensevoice/bin/activate- 安装依赖包
pip install https://github.com/AXERA-TECH/pyaxengine/releases/download/0.1.3.rc2/axengine-0.1.3-py3-none-any.whl
pip install -r requirements.txt- 运行
python main.py -i test.mp3| 参数名称 | 说明 | 默认值 |
|---|---|---|
| --input/-i | 输入音频文件 | - |
| --language/-l | 识别语言，支持 auto, zh, en, yue, ja, ko | auto |
说明
首次运行会自动下载模型。
运行结果如下：
(sensevoice) m5stack@raspberrypi:~/rsp/SenseVoice $ python main.py -i test_en.mp3
[INFO] Available providers: ['AXCLRTExecutionProvider']
input_audio: test_en.mp3
language: auto
use_itn: True
model_path: /home/m5stack/rsp/SenseVoice/models/SenseVoice/sensevoice_ax650/sensevoice.axmodel
[INFO] Using provider: AXCLRTExecutionProvider
[INFO] SOC Name: AX650N
[INFO] VNPU type: VNPUType.DISABLED
[INFO] Compiler version: 4.0 156de6f7
RTF: 0.015400904924311537 Latency: 0.463259220123291s Total length: 30.08s
['You want to be a nurse or an archi', "A lawyer or a member of our military. you''re going to need a", 'Eduducation for every single one of those caree', 'Not drop out of school and just drop into a good j', "You''ve got to train for it", "And learn for it. And this isn't just impo", 'lifeIn your own future. what you make', 'Will decide nothing less than the future of this country..']