huggingface_hub

f60aef03 - [Inference Providers] deepinfra: add text-to-speech support (#4559)

[Inference Providers] deepinfra: add text-to-speech support (#4559) Register DeepInfraTextToSpeechTask on the deepinfra provider, targeting the OpenAI-compatible `POST /v1/openai/audio/speech` endpoint and returning the raw audio bytes. The mapped provider model is applied last so caller parameters cannot override it; `voice` is passed through untouched. Co-authored-by: Oguz Vuruskaner <oguz@deepinfra.com> Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Co-authored-by: célina <hanouticelina@gmail.com>