senseaudio-voice

What This Skill Does

The senseaudio-voice skill provides a robust, multi-modal voice processing engine for the OpenClaw AI agent. It seamlessly integrates both Text-to-Speech (TTS) and Automatic Speech Recognition (ASR) capabilities into a single, cohesive workflow. By intelligently switching between the high-quality SenseAudio API (best for Chinese synthesis) and Microsoft's Edge TTS (ideal for English and Japanese), this skill ensures a natural-sounding experience regardless of the target language.

Key features include:

• Automated Language Intelligence: The skill automatically detects the input language and chooses the optimal engine, reducing the need for manual configuration.
• WAV/MP3 Export: Provides flexible file output, with audio files automatically organized by date in your workspace directory.
• Flexible Synthesis: Support for multiple voices, adjustable speed, and variable volume settings, allowing for personalized AI persona behavior.
• Robust Playback: Built-in support for automatic playback after generation, ensuring immediate feedback for user queries.

Installation

To install this skill, run the following command in your terminal:

clawhub install openclaw/skills/skills/cowboy231/kid-point-voice-component

Once installed, ensure your ~/.openclaw/openclaw.json configuration file is updated with your SENSE_API_KEY if you intend to use the premium SenseAudio features for Chinese language synthesis.

Use Cases

This skill is designed for interactive scenarios that require human-like response. Common use cases include:

• Educational Tools: Using the default child-voice to create engaging study assistants for children.
• Internationalized Agents: Building agents that communicate fluently across Chinese, English, and Japanese markets without needing separate deployments.
• Accessibility: Converting textual AI outputs into spoken audio for users who prefer auditory information.
• Workflow Automation: Creating automated reminders or status notifications that can be played on a local speaker via system commands.

Example Prompts

1. "Speak the following in a friendly voice: 'Good morning, please review your dashboard for today's tasks.'"
2. "Convert this text to Japanese using the female voice: 'おはようございます、今日の予定を確認しましょう。'"
3. "Summarize the last report and play the audio using the male_0004_a voice profile."

Tips & Limitations

• Chinese Language: For optimal results, ensure your SenseAudio credentials are valid. This provides access to natural-sounding Chinese voices that are superior to the fallback options.
• Edge TTS: For English and Japanese, no API key is required as it connects directly to Microsoft's reliable edge service.
• Storage Management: Files are saved in {workspace}/audio/YYYY-MM-DD/. Periodically clear this folder to save disk space if the agent is used for high-volume tasks.
• Dependency: The skill relies on the requests library. Ensure your environment has basic internet access to reach the external TTS endpoints.