← Back to Skills Marketplace
266
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install polyphone
Description
Fix Chinese polyphone (多音字) mispronunciation in TTS by auto-detecting ambiguous characters and applying pinyin annotations. Use when users complain about wro...
Usage Guidance
This skill appears coherent and does what it says, but it sends the text you provide to a third‑party API at senseaudio.cn. Do not send sensitive secrets or private data to the skill unless you trust that service and its API key handling. Make sure you store the SENSEAUDIO_API_KEY securely (use a least‑privilege key and rotate it if possible). Confirm the cloned voice requirement before running (dictionary only works with cloned voices and model SenseAudio-TTS-1.5). Ensure curl, jq, and xxd are available on the host. If privacy is a concern, test with non-sensitive examples and review SenseAudio's privacy/terms before using real user data.
Capability Analysis
Type: OpenClaw Skill
Name: polyphone
Version: 1.0.0
The skill provides a utility for correcting Chinese polyphone pronunciation using the SenseAudio API (api.senseaudio.cn). It is classified as suspicious due to a potential shell injection vulnerability in the SKILL.md file, where user-controlled text and dictionary arrays are inserted directly into a single-quoted curl command template. While the logic is aligned with the stated purpose and no malicious intent was found, the lack of input sanitization in the provided shell instructions poses a security risk. The skill also utilizes xxd for hex-to-binary conversion of API responses, which is a powerful utility but appears to be used legitimately here.
Capability Assessment
Purpose & Capability
Name/description (polyphone fixes for TTS) align with required items: SENSEAUDIO_API_KEY and CLI tools (curl, jq, xxd) are appropriate for calling SenseAudio's API and decoding returned audio. Requiring a cloned voice and the specific model is documented and consistent.
Instruction Scope
SKILL.md stays on task: scanning text for ambiguous characters, prompting for confirmation, building a dictionary, and calling https://api.senseaudio.cn. It instructs saving response.json and decoding audio. Important privacy note: the skill will send user-provided text (potentially sensitive) to senseaudio.cn — this is expected for a remote TTS but worth flagging.
Install Mechanism
No install spec (instruction-only) so nothing is downloaded or written by the skill itself. Required binaries are standard CLI tools and reasonable for the demonstrated curl/jq/xxd workflow.
Credentials
Only one required environment variable (SENSEAUDIO_API_KEY) is declared and used as the bearer token for the SenseAudio API. That is proportionate to the skill's function; no unrelated secrets or config paths are requested.
Persistence & Privilege
always:false and no instructions to modify other skills or system-wide settings. The skill writes/reads a local response.json as part of normal operation, which is consistent with its purpose.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install polyphone - After installation, invoke the skill by name or use
/polyphone - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of polyphone-aware Chinese TTS pronunciation control.
- Automatically detects ambiguous polyphonic Chinese characters (多音字) in user text.
- Suggests likely pronunciations using context clues, and prompts the user to confirm or adjust them.
- Constructs a pinyin-annotated `dictionary` array to ensure accurate pronunciation in TTS synthesis.
- Requires a cloned voice and the SenseAudio-TTS-1.5 model for dictionary support; reminds users if missing.
- Allows iterative correction: update dictionary entries and re-synthesize as needed.
- Reports applied dictionary items, output file path, duration, and character count after synthesis.
Metadata
Frequently Asked Questions
What is Polyphone TTS?
Fix Chinese polyphone (多音字) mispronunciation in TTS by auto-detecting ambiguous characters and applying pinyin annotations. Use when users complain about wro... It is an AI Agent Skill for Claude Code / OpenClaw, with 266 downloads so far.
How do I install Polyphone TTS?
Run "/install polyphone" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Polyphone TTS free?
Yes, Polyphone TTS is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Polyphone TTS support?
Polyphone TTS is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Polyphone TTS?
It is built and maintained by scikkk (@scikkk); the current version is v1.0.0.
More Skills