docs(openrouter-tts): sync with openrouter-web changes - #153
docs(openrouter-tts): sync with openrouter-web changes#153devin-ai-integration[bot] wants to merge 3 commits into
Conversation
Original prompt from Automation
|
🤖 Devin AI EngineerI'll be helping with this pull request! Here's what you should know: ✅ I will automatically:
Note: I can only respond to comments from users who have write access to this repository. ⚙️ Control Options:
|
Summary
Syncs
openrouter-ttswith openrouter-web #31562, which added stateless voice cloning to the public/api/v1/audio/speechrequest schema.New public request field, documented in the parameter table plus a worked example:
input_references: [ { type: "input_audio", input_audio: { data: "<base64 | data URI>", format?: "wav" } }, { type: "text", text: "<transcript of the reference audio>" } // optional, at most one ]Server-side constraints the skill now states: exactly one
input_audiopart is required, at most onetexttranscript part, and reference audio is capped at 20 MiB of base64 (15 MiB decoded).Two corrections that follow from the same change:
supports_voice_cloningon the public endpoint projection), not a models-list field, so discovery routes throughGET /api/v1/models/{author}/{slug}/endpoints. Requests carrying reference audio are only routed to endpoints where it is true.voiceis optional at the schema level, not unconditionally required; the model/provider-specific requirement is kept as a caveat.No other skill needed an update from the last 24 hours of merges — the remaining candidates were provider-adapter and serialization changes, internal billing/pricing SKUs,
cfw-frontend-apiroutes, or public surfaces no existing skill teaches.Link to Devin session: https://openrouter.devinenterprise.com/sessions/fc599ee099b64d18af727a012bf0db81