feishu-edge-tts

What This Skill Does

The feishu-edge-tts skill leverages Microsoft's advanced Edge Text-to-Speech (TTS) engine to convert text into high-quality, natural-sounding audio files. It integrates directly with Feishu (Lark) to send these audio clips as native voice messages. This solution is completely free, bypassing the need for expensive API keys while maintaining the professional audio quality expected of Azure-powered services. It provides deep customization options, including granular control over speaking rates and pitch adjustments, making it ideal for creating expressive and context-aware notifications.

Installation

To begin, ensure your system has Python 3.x installed. First, install the edge-tts library via pip: pip install edge-tts. Next, you must install FFmpeg, a multimedia framework required to encode the audio into the correct format for Feishu; use sudo yum install -y ffmpeg on CentOS or sudo apt-get install -y ffmpeg on Ubuntu. Finally, clone the skill repository from openclaw/skills and configure your environment variables: FEISHU_APP_ID, FEISHU_APP_SECRET, and FEISHU_CHAT_ID within your environment or the local config.sh file to authenticate with your Feishu application.

Use Cases

This skill is perfect for automated notification systems, such as alerting team members to server statuses, CI/CD pipeline results, or daily schedule updates. It is also excellent for accessibility, allowing users who prefer listening to messages rather than reading them to stay informed. Furthermore, developers can use it to build interactive "talking" agents that provide feedback with diverse personas, such as using warm, friendly voices for greetings or professional, clear voices for critical error reporting.

Example Prompts

1. "Send a morning greeting to the team chat using the zh-CN-XiaoxiaoNeural voice that says: Good morning everyone, let's start the day with focus!"
2. "Use the en-US-JennyNeural voice to announce that the production deployment has been completed successfully."
3. "Send a reminder in the project group: Please remember to submit your weekly status reports by 5 PM today, using the YunxiNeural voice."

Tips & Limitations

To get the best performance, experiment with the pitch and rate settings to match your desired tone; subtle adjustments (e.g., rate 1.1) can often make AI voices feel more natural. Remember that this skill requires an active internet connection to communicate with the Microsoft Edge TTS service. Since it sends native voice files, ensure your Feishu bot has permission to upload files to the target chat. If you need to generate voices offline or for high-frequency mass distribution, consider caching frequently used phrases to reduce latency.