Speech-to-Text (ASR)

ASR (Automatic Speech Recognition) converts spoken language into text in real time. It is a foundation for voice assistants, meeting notes, intelligent customer service, and voice input.

Hardware	Model	Performance benchmark
RDK X5 USB speaker / wired headset	SenseVoice	RTF (inference_time_seconds/audio_seconds): 0.52 (8-core CPU)

Hardware Connection

• Plug a USB speaker into a USB port on the RDK X5 development board.
• Plug a wired headset into the earphone audio jack on the RDK X5 development board.

Environment Setup

Install Dependencies

sudo apt update
sudo apt install libasound2-dev

Get the Sample Code and Build

wget https://archive.d-robotics.cc/downloads/rdk_demo/rdk_x5_demo/asr_demo_cpp.tar.gz
tar zxvf asr_demo_cpp.tar.gz
cd asr_demo_cpp
cmake -S . -B build
cmake --build build -j6

Run the Sample

Tip

• This feature captures microphone input for ASR. Try it in a quiet environment. A noise-canceling audio device is recommended.
• To specify the audio device at runtime, see Earphone Audio Jack. For more configuration, see the README in the sample code.
• If ASR cannot recognize speech, try lowering threshold and vad-threshold.

# Run from the asr_demo_cpp directory
./build/asr_microphone --device plughw:0,0 --threshold 0.03 --vad-threshold 0.5