← Back to Skills Marketplace
312
Downloads
0
Stars
0
Active Installs
5
Versions
Install in OpenClaw
/install u2-tts
Description
Text-to-speech conversion using UniSound's TTS WebSocket API for generating high-quality Chinese Mandarin audio from text. Supports multiple voices, adjustab...
Usage Guidance
This skill appears to do exactly what it claims: run the included Python script which opens a WebSocket to UniSound's TTS service and saves audio to results/. Before installing/running: (1) verify you trust the UNISOUND_APPKEY/UNISOUND_SECRET you supply (do not use credentials you don't control), (2) inspect scripts/tts.py (it is short and readable) if you have privacy concerns about sending text to an external API, and (3) be aware the script will open outbound network connections to wss://ws-stts.hivoice.cn and write files to a local results/ folder. Also note the README contains example/demo credentials—do not reuse those for production.
Capability Analysis
Type: OpenClaw Skill
Name: u2-tts
Version: 1.0.5
The u2-tts skill is a legitimate implementation of a text-to-speech client for the UniSound WebSocket API. The core logic in `scripts/tts.py` handles authentication via SHA256 signatures and manages binary audio streams as expected for this service. While `SKILL.md` contains highly restrictive instructions for the AI agent (e.g., 'MANDATORY RESTRICTIONS' forbidding fallback methods) and provides hardcoded test credentials, these are accompanied by explicit security warnings regarding data privacy and the intended use of test keys. No evidence of malicious data exfiltration, unauthorized execution, or obfuscation was found.
Capability Assessment
Purpose & Capability
The name/description, required env vars (UNISOUND_APPKEY, UNISOUND_SECRET), required binary (python), and included script all align with a UniSound TTS client. There are no unrelated credentials or binaries requested.
Instruction Scope
SKILL.md instructs the agent to only run the included Python script and not to perform local synthesis or fallbacks. That is restrictive but consistent with a remote-API-only design. SKILL.md also contains example/demo credentials embedded in the docs—these are likely test keys and should not be treated as production secrets.
Install Mechanism
This is an instruction-only skill with a bundled Python script and a requirements.txt. No external arbitrary downloads or install hooks are present. Dependencies are standard (websocket-client, optional gevent).
Credentials
Only two env vars are required (appkey and secret), which matches the TTS API's needs. The script reads only those env vars (and respects CLI overrides). The primary credential is the UNISOUND_SECRET as declared.
Persistence & Privilege
always is false and the skill does not request elevated or persistent system privileges. It writes output files to a local results/ directory and has a .claude settings file granting Bash/python execution permissions (expected for running the script). It does not modify other skills or system configs.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install u2-tts - After installation, invoke the skill by name or use
/u2-tts - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.5
u2-tts v1.0.5
- Updated documentation for simplified configuration, emphasizing environment variable setup for credentials.
- Added support for command-line credential override via `--appkey` and `--secret` arguments.
- Default sample rate changed to 24kHz for higher output quality.
- Clarified supported parameter options and output file formats.
- Improved step-by-step usage examples and troubleshooting guidance for first-time users.
v1.0.4
u2-tts v1.0.4 changelog:
- Added built-in test credentials for quick evaluation, enabling immediate use without configuration.
- Updated documentation to clarify the distinction between test and production credentials, including usage and security warnings.
- Configuration via environment variables is still supported and recommended for production.
- No code changes detected; documentation update only.
v1.0.2
- Documentation updated to focus on secure configuration, usage guidelines, and API credential setup.
- The sample README.md file was removed from the project.
- Enhanced instructions for safe environment variable configuration (.env, export) with warnings against sharing credentials.
- Emphasized usage restrictions: only use UniSound TTS API via the provided script, no local or fallback methods.
- Clarified error handling, sample commands, supported voices, and parameter adjustment for text-to-speech use cases.
v1.0.1
u2-tts 1.0.1 changelog:
- Replaced the old readme.txt file with a new, detailed README.md.
- Updated documentation to clarify usage via Python API and command line, featuring new example commands and scripts paths.
- Included actual demo UniSound API credentials in documentation for quick start (replace with your own for production).
- All usage examples and installation instructions now reference the updated scripts folder and naming conventions.
v1.0.0
UniSound TTS (U2-TTS) 1.0.0
- Initial release with WebSocket streaming TTS support for Mandarin Chinese.
- Supports multiple voices (e.g., xiaofeng, xiaoyan) and adjustable speech parameters (speed, volume, pitch, brightness).
- Allows output in mp3, wav, and pcm formats, with sample rates of 8k or 16k.
- Authentication via SHA256 signature; credentials required via environment variables.
- Includes usage examples, error handling, and long text/audiobook conversion.
- Requires Python 3.6+ and the websocket-client library.
Metadata
Frequently Asked Questions
What is U2-tts?
Text-to-speech conversion using UniSound's TTS WebSocket API for generating high-quality Chinese Mandarin audio from text. Supports multiple voices, adjustab... It is an AI Agent Skill for Claude Code / OpenClaw, with 312 downloads so far.
How do I install U2-tts?
Run "/install u2-tts" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is U2-tts free?
Yes, U2-tts is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does U2-tts support?
U2-tts is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created U2-tts?
It is built and maintained by aaiccee (@aaiccee); the current version is v1.0.5.
More Skills