{"skill":{"slug":"local-tts-workflow","displayName":"Local Tts Workflow","summary":"OpenClaw text-to-speech workflow for an OpenAI-compatible TTS server, including remote/self-hosted deployments such as vLLM Omni. Use when configuring, testi...","tags":{"latest":"1.0.1"},"stats":{"comments":0,"downloads":116,"installsAllTime":1,"installsCurrent":1,"stars":0,"versions":2},"createdAt":1775032280683,"updatedAt":1775224007478},"latestVersion":{"version":"1.0.1","createdAt":1775223729462,"changelog":"- Updated server references from a remote IP to `127.0.0.1` to assume local deployment.\n- Noted that models may be loaded via local file paths (e.g., `./models/qwen3-tts-0.6b-mlx`), not just registry pulls, and clarified OpenClaw model detection flow accordingly.\n- Documented the local OpenAPI schema endpoint (`/openapi.json`) as the reference source for API details; clarified distinction between schema and server health.\n- Revised TTS endpoint examples to match new field names (`response_format` instead of `format`, added `streaming_interval`), and updated streaming/non-streaming default behaviors per current OpenAPI spec.\n- Expanded documentation of optional and required API parameters and urged users to consult the local OpenAPI for non-parity with OpenAI’s API.\n- Added clarifications for troubleshooting and debugging patterns, focusing on local-path deployments and real server vs. integration failures.","license":"MIT-0"},"metadata":null,"owner":{"handle":"mozi1924","userId":"s174c63w5njkpvrfq9a8zxbj5983hwcp","displayName":"Mozi Arasaka","image":"https://avatars.githubusercontent.com/u/70095417?v=4"},"moderation":null}