BUG: Gemini 3.8 Live API overrides language_code / ignores system prompt accent formatting

Gemini API
1

Model Impacted: gemini-3.8-live (Regression from gemini-3.1-flash-live-preview)

Environment: Pipecat / WebSockets API integration

Description:
In previous preview versions (gemini-3.1), passing a non-US language code or system prompt instructions safely forced the Live API output voices to speak English with a regional accent (e.g., British or Spanish) while maintaining clean English text tokens.

With the stable release of gemini-3.8-live, this behavior is entirely broken/unstable:

  1. Passing a language_code to force an accent layer causes the audio generation weights to aggressively drift. The model will begin speaking the target language (e.g., Spanish) even when processing correct English text tokens.
  2. Traditional text-based system instructions to adopt an accent are ignored, defaulting the voice profiles to General American phonetics.

Impact on Use Case:
This breaks applications that require clean English spoken alongside other languages with an accented voice profile.

Steps to Reproduce:
Pass an English string to gemini-3.8-live with language_code set to a non-US variant. The model will either drift into speaking that language natively or fail to shift vocal phonetics away from the American base profile. Or pass clear instructions in the prompt and there is no accent change.

2

Ya 3.8 live is …, it does not process any accents. I went back to 2.5.

3

Just adding to this. I’m facing exactly this problem, and 3.8 is simply impossible when it comes to having it follow a certain voice profile, where 3.1 performed this task perfectly. Unfortunately I’m switching back to 3.1 until this major regression is resolved.

4

The problem. 2.5 is retired 13 December and people need time to migrate. I don’t know who is running the project or the QA side but my stress levels are going through the roof and i am sure people running with 2.5 and 3.8 are suffering financial harm.

My recommendation to google is suspend all costs on audio output tokens on 3.8 live until its resolved plus a switch over grace period. Our job is to build solutions with trusted technology. Not hit brick walls and end up being charged extra even if a hybrid interim solution is possible (another separate tts). To me this is a P1 because the platform is fundamentally broken.

5

Same frustration we are going through tried all three live models 2.5 , 3.1 preview, 3.8 . None of them are stable to run in PROD.