aliyun-asr
by jixsonwang
CommunicationBrief
Pure Aliyun ASR skill for voice message transcription, supports multiple channels including Feishu.
Description
What This Skill Does
Lightweight skill for transcribing voice messages using Alibaba Cloud's NLS (Natural Language Service) ASR API. Converts incoming audio to text and passes it to the AI as a regular text message. No text-to-speech or voice reply generation is involved.
Keeps the skill surface minimal by handling only ASR, avoiding the complexity and cost of a combined speech skill that also generates voice replies.
When to Use It
- Transcribing Feishu voice messages into text for AI processing
- Handling WhatsApp voice notes in an OpenClaw agent workflow
- Converting OGG or OPUS audio from Telegram voice messages
- Processing AMR audio from mobile messaging apps
- Enabling voice input for users who prefer speaking over typing
Example Workflow
User asks: sends a voice message via Feishu
- 1
OpenClaw receives the audio file attachment
- 2
Skill calls Alibaba Cloud NLS ASR API with the audio data
- 3
API returns a text transcript
- 4
Skill passes the transcript as a plain text user message to the AI
- 5
AI responds with a text reply
The spoken message is processed as if the user had typed it
Requirements
- Alibaba Cloud account with NLS (智能语音交互) service enabled
- Alibaba Cloud RAM sub-account access key ID and secret with AliyunNLSFullAccess permission
- Alibaba Cloud NLS AppKey
- Configuration file at /root/.openclaw/aliyun-asr-config.json with credentials
Installation
clawhub install jixsonwang/aliyun-asrRun the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.