Summary
The /tts (text-to-speech) endpoint is documented at https://openrouter.ai/docs/api/api-reference/tts/create-tts and exists in the OpenAPI spec (added in #141), but no corresponding SDK module was generated. There is no src/openrouter/tts.py and no tts entry in the OpenRouter client.
The same gap exists in the Go SDK (OpenRouterTeam/go-sdk) — no tts.go was generated there either.
Likely cause
Speakeasy may need a config adjustment to handle application/octet-stream binary responses, which differ from the JSON responses all other endpoints return. This may be why it was silently skipped during generation in both SDKs.
Expected behavior
client.tts.create_speech(model=..., input=..., voice=...) returning bytes, following the same sync/async pattern as other resources like video_generation.
Summary
The
/tts(text-to-speech) endpoint is documented at https://openrouter.ai/docs/api/api-reference/tts/create-tts and exists in the OpenAPI spec (added in #141), but no corresponding SDK module was generated. There is nosrc/openrouter/tts.pyand nottsentry in theOpenRouterclient.The same gap exists in the Go SDK (
OpenRouterTeam/go-sdk) — notts.gowas generated there either.Likely cause
Speakeasy may need a config adjustment to handle
application/octet-streambinary responses, which differ from the JSON responses all other endpoints return. This may be why it was silently skipped during generation in both SDKs.Expected behavior
client.tts.create_speech(model=..., input=..., voice=...)returningbytes, following the same sync/async pattern as other resources likevideo_generation.