feishu-voice-assistant

What This Skill Does

The feishu-voice-assistant is a specialized skill for the OpenClaw AI ecosystem that bridges the gap between text-based AI processing and native audio communication within the Feishu (Lark) collaboration platform. By integrating the Duby TTS (Text-to-Speech) engine, this skill converts any arbitrary text input into high-quality, natural-sounding audio files. Once generated, the skill handles the authentication and file transfer protocols required to deliver these audio snippets directly into Feishu chats as native voice messages. This allows users to leverage AI-generated responses that are more personal, expressive, and easier to consume on the go than standard text messages.

Installation

To integrate this skill into your environment, use the OpenClaw installer. Execute the following command in your terminal: clawhub install openclaw/skills/skills/autogame-17/feishu-voice-assistant

Before running, ensure your .env file is properly configured with your DUBY_API_KEY for speech generation and the necessary Feishu credentials (App ID and App Secret) to permit file uploads and message delivery to your workspace.

Use Cases

This skill is ideal for teams and individuals looking to modernize their notification systems. Common use cases include: 1. Automated voice alerts for urgent server monitoring statuses. 2. Personalized AI "voice notes" for team updates in project channels. 3. Accessibility-focused messaging, ensuring teammates can listen to updates instead of reading long-form text. 4. Customer support automation where a human-like voice response is preferred over a robotic text notification.

Example Prompts

1. "Use the feishu-voice-assistant to send a message to the engineering team channel saying: The production deployment has finished successfully."
2. "Send a voice note to user ou_12345 telling them: Reminder, we have a sync meeting in 10 minutes, please prepare your status report."
3. "Convert this text to speech using the Xinduo voice and send it to the product feedback group: We have reviewed the latest UI changes and they look great."

Tips & Limitations

To get the best results, keep messages concise; very long strings may hit character limits imposed by the TTS provider. Currently, the skill supports the default 'Xinduo' voice, though you can specify alternative Duby Voice IDs via the --voice flag to match specific brand tones. Note that this skill requires an active internet connection to communicate with both the Duby API and Feishu servers. Ensure your Feishu bot has the 'Send Message' and 'Upload File' permissions enabled in the Developer Console to avoid authorization errors.