OpenTypeless

Aliyun Qwen3-ASR

Qwen3-ASR is integrated with DashScope real-time WebSocket protocol and duplex transcription.

Official Documentation

DashScope Real-time Speech Recognition

Official feature and model overview for real-time ASR.

https://help.aliyun.com/zh/model-studio/realtime-speech-recognition

DashScope WebSocket Protocol

WebSocket message protocol and run-task flow.

https://help.aliyun.com/zh/model-studio/websocket-for-paraformer-real-time-service

Aliyun Model Studio API Key

Official API key creation guide.

https://help.aliyun.com/zh/model-studio/get-api-key

Get API Key

  1. Open Aliyun Model Studio / DashScope and enable ASR service.
  2. Create API key with speech permissions.
  3. Store key securely.

Configure in OpenTypeless

  1. Settings > Speech Recognition > Provider: Qwen3-ASR (Aliyun).
  2. Paste key and run Test.
  3. If auth fails, verify service entitlement and account scope.

Recommended Configuration

OpenTypeless uses DashScope duplex WebSocket with PCM16 audio and intermediate results.

Provider: aliyun-qwen3-asr
Protocol: WebSocket
Endpoint: wss://dashscope.aliyuncs.com/api-ws/v1/inference
Auth header: Authorization: Bearer <DASHSCOPE_API_KEY>
Audio: pcm, 16000 Hz, mono
Task actions: run-task / finish-task

Important

⚠️
A valid key without real-time ASR entitlement can still return authorization failures.