Zhipu GLM-ASR

GLM-ASR is integrated through Zhipu BigModel API with OpenAI-like request shape.

Official Documentation

Zhipu Speech Transcription API

Official GLM speech transcription reference.

https://docs.bigmodel.cn/api-reference/%E6%A8%A1%E5%9E%8B-api/%E8%AF%AD%E9%9F%B3%E8%BD%AC%E5%86%99

Zhipu API Key Management

Project key creation and rotation.

https://open.bigmodel.cn/usercenter/proj-mgmt/apikeys

Get API Key

Create a BigModel project.
Generate project API key.
Verify project quota/model entitlement.

Configure in OpenTypeless

Settings > Speech Recognition > Provider: GLM-ASR.
Paste key and run Test.
Allow outbound HTTPS to open.bigmodel.cn in restricted environments.

Recommended Configuration

OpenTypeless sends WAV transcription requests to Zhipu audio endpoint.

Provider: glm-asr
Protocol: HTTPS multipart upload
Endpoint: https://open.bigmodel.cn/api/paas/v4/audio/transcriptions
Auth header: Authorization: Bearer <ZHIPU_API_KEY>
Model: glm-asr-2512
Extra field: stream=false

Operational Tip

💡

Keep STT output raw if needed, and rely on LLM polish for punctuation/style normalization.