Real Time Whisper On NVIDIA Jetson Nano

This is a demo of real time speech to text with OpenAI's Whisper model. It works on NVIDIA Jetson Nano and uses the microphone and speaker on the NVIDIA Jetson Nano Developer Kit.

Hardware Requirements

• reComputer (Or other devices based on Jetson)
• reSpeaker (Or other USB interface microphones)
• Monitor, mouse, and keyboard

To install dependencies simply run

git clone https://github.com/LJ-Hao/Deploy-Whisper-on-NVIDIA-Jetson-Orin-for-Real-time-Speech-to-Text.git
cd Deploy-Whisper-on-NVIDIA-Jetson-Orin-for-Real-time-Speech-to-Text
sudo apt install python3-pyaudio
pip3 install -r requirements.txt
sudo apt update && sudo apt install ffmpeg

And you can use following command to check wheather the dependencies install successfully.

python test.py

If you can see the following result, it means that the dependencies are installed successfully.

Run this command to set the microphone rate

arecord -D hw:2,0 --dump-hw-params

Run this command to start the demo

python3 main.py

Result

For more information on Whisper please see https://github.com/openai/whisper