does TTS v3 support streaming in any way?

[steel tangle]

AI-generated Answer for betterhomes01_07039

Yes. Eleven v3 supports HTTP streaming (chunked transfer) via:

• Text-to-Speech: POST /v1/text-to-speech/:voice_id/stream (set model_id: "eleven_v3"). (elevenlabs.io)
• Text-to-Dialogue: POST /v1/text-to-dialogue/stream (defaults model_id to "eleven_v3"). (elevenlabs.io)
• Streaming uses incremental audio over HTTP; see streaming overview. (elevenlabs.io)

However, WebSocket/input streaming is not available for the eleven_v3 model. (elevenlabs.io)

Note: v3 has higher latency and isn’t suited for real-time conversational use; use v2/v2.5 Turbo/Flash for realtime. (help.elevenlabs.io)

Sources:
Stream speech | ElevenLabs Documentation
Stream dialogue | ElevenLabs Documentation
Streaming | ElevenLabs Documentation
Generate audio in real-time | ElevenLabs Documentation
What is Eleven v3? – ElevenLabs

[humble escarp]

@quartz wing

[steel tangle]

AI-generated Answer for betterhomes01_07039

Yes.

• HTTP streaming: Supported via the Text to Speech “Stream speech” endpoint (POST /v1/text-to-speech/:voice_id/stream) and Text to Dialogue “Stream dialogue” endpoint (POST /v1/text-to-dialogue/stream). Use model_id=eleven_v3. (elevenlabs.io)
• WebSocket streaming: Not supported for the eleven_v3 model (use Flash/Turbo for realtime WS). (elevenlabs.io)

Sources:
Stream speech | ElevenLabs Documentation
Generate audio in real-time | ElevenLabs Documentation

[quartz wing]

thanks!

[steel tangle]

AI-generated Answer for tofu_tm

You’re welcome! If you need anything else with ElevenLabs, just let me know.

[humble escarp]

@glass quest