tts-whatsapp

What This Skill Does

The tts-whatsapp skill is a powerful automation tool that integrates text-to-speech (TTS) capabilities directly with WhatsApp messaging. It leverages the Piper TTS engine to generate high-quality, natural-sounding voice messages from text input. Once the audio is generated, the skill automatically transcodes the file into the OGG/Opus format required by WhatsApp and utilizes the Clawdbot agent to deliver the message to any individual contact or group. This tool eliminates the need for manual voice recording, allowing users to broadcast information, updates, or personal messages in over 40 languages with extreme speed and efficiency.

Installation

To get started, ensure you have the necessary system dependencies. First, install Piper TTS via pip3 install --user piper-tts. Next, install FFmpeg, which is required for audio format conversion—use brew install ffmpeg on macOS or apt install ffmpeg on Linux. Download your preferred voice models from the official Rhasspy Hugging Face repository and move them to ~/.clawdbot/skills/piper-tts/models/. Finally, install the skill itself by running the command clawhub install openclaw/skills/skills/hopyky/tts-whatsapp in your terminal.

Use Cases

This skill is perfect for scenarios where a human touch is needed without the effort of real-time recording. Businesses can use it for automated status notifications or appointment reminders in the customer's native language. Content creators can quickly narrate scripts for group updates. It is also highly effective for accessibility, enabling users to send messages in languages they may not speak fluently, or for users who prefer listening to messages rather than reading them in high-noise environments.

Example Prompts

1. "tts-whatsapp 'Your scheduled maintenance is confirmed for tomorrow at 10 AM' --target '+447700900123'"
2. "tts-whatsapp 'Bonjour, voici le compte rendu de la réunion' --lang 'fr_FR' --voice 'siwis' --target '[email protected]'"
3. "tts-whatsapp 'The server update is complete, everything is back online' --target '+15550199' --quality 'high'"

Tips & Limitations

To maintain performance, keep messages concise; the system averages around 2-3 seconds for a full delivery cycle. If you are sending to a large group, ensure your group ID is correctly retrieved. Note that the quality of the audio is highly dependent on the voice model chosen—'high' quality settings will increase generation time slightly but provide superior clarity. Always ensure your environment variables in clawdbot.json are set correctly to avoid repetitive typing of common target numbers.