gemini-3.8-live: replicated_voice_config is accepted but ignored on the Gemini Developer API (API key)

Hi,

We are building a phone voice agent (Uzbek and Russian) on gemini-3.8-live through the Gemini Developer API (AI Studio API key, not Vertex AI). We need the agent to speak with our own brand voice.

What we do:

  • google-genai 2.30.1, client.aio.live.connect(model=“gemini-3.8-live”, …)
  • speech_config.voice_config.replicated_voice_config with mime_type=“audio/wav” and voice_sample_audio = a 16-second, 24 kHz, 16-bit mono WAV of clean speech.

What happens:

  • The session connects with no error and the model answers with audio.
  • But the output voice does not change. It sounds like the default prebuilt voice (close to “Kore”).
  • A white-noise sample is also accepted without any error, so the reference does not seem to be validated or used.
  • A/B test with the same prompt: Kore = 217 Hz median F0, Puck = 103 Hz, our replicated voice = 181–226 Hz, a deliberately lower-pitched reference (163 Hz) still gave 211 Hz. So the reference does not steer the voice.

Questions:

  1. Is replicated_voice_config for gemini-3.8-live supported on the Gemini Developer API, or only for allowlisted Vertex AI customers?
  2. If access is needed, how can we request it?
  3. Can a voice created with POST /v1beta/voices (voice_… id) be used with gemini-3.8-live? The docs list only gemini-3.8-flash-tts and gemini-3.8-flash-lite-tts.
  4. Is Uzbek (uz-UZ) planned as a consent locale for voice replication?

Thank you!

Hello @Ilkhom ,

Thank you for the detailed report. We are looking into the replicated_voice_config behavior on gemini-3.8-live. To answer your questions in the meantime:

  1. Custom voice replication (replicated_voice_config) and stored custom voices created via POST /v1beta/voices (voice_... IDs) are currently supported on the TTS models (gemini-3.8-flash-tts and gemini-3.8-flash-lite-tts) and are not supported on gemini-3.8-live in the Gemini Developer API.
  2. If you would like us to keep track of your project as we look into this, please feel free to share your Google Cloud Project ID via direct message.