Skip to content

fix(tts): resolve Gemini TTS models from catalog - #2055

Closed
nguyenha935 wants to merge 1 commit into
decolua:masterfrom
nguyenha935:fix/gemini-tts-model-resolution
Closed

nguyenha935 wants to merge 1 commit into
decolua:masterfrom
nguyenha935:fix/gemini-tts-model-resolution

Conversation

@nguyenha935

Copy link
Copy Markdown
Contributor

Summary

  • Fix Gemini TTS requests resolving to models/undefined:generateContent when the handler has no ttsConfig.models.
  • Resolve Gemini TTS models from the shared TTS catalog and provider registry, with a safe fallback.
  • Add gemini-3.1-flash-tts-preview to Gemini TTS catalogs and cover model/voice parsing with unit tests.

Tests

  • tests/node_modules/.bin/vitest run tests/unit/gemini-tts.test.js --config tests/vitest.config.js
  • tests/node_modules/.bin/vitest run tests/unit/minimax-tts.test.js --config tests/vitest.config.js
  • tests/node_modules/.bin/vitest run tests/unit/provider-display-split.test.js --config tests/vitest.config.js

@decolua

decolua commented Jun 26, 2026

Copy link
Copy Markdown
Owner

Thanks @nguyenha935! Merged into master. 🙏

@decolua decolua closed this Jun 26, 2026
diegosouzapw pushed a commit to cheaper-inference/OmniRoute that referenced this pull request Jun 26, 2026
Add gemini-3.1-flash-tts-preview as the first (default) entry in the
Vertex TTS catalog in audioRegistry.ts, and register it with kind:"tts"
in the Gemini provider registry. This makes the new model the default
when the caller supplies a bare voice name, and keeps model resolution
catalog-driven so future TTS additions are automatically available.

Co-authored-by: nguyenha935 <208228297+nguyenha935@users.noreply.github.com>
Inspired-by: decolua/9router#2055
@nguyenha935
nguyenha935 deleted the fix/gemini-tts-model-resolution branch July 1, 2026 19:23
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants