Platform edition
Gemini for Windows
Google Gemini cloud transcription with live text, plus Gemini and Gemma text processing with separate model catalogs.
- Version
- 1.3.10
- Plugin ID
- com.
typewhisper. gemini - Requirements
- TypeWhisper 1.1.0 or newer · Windows 10 or newer · A Gemini API key from Google AI Studio
Inside this edition
Features in this edition
- Recorded-audio transcription with Gemini 3.5 Transcribe or another transcribe model from your catalog
- Live text while recording with transcription models that have a live version
- Smart and Verbatim transcription modes, language hints, and dictionary terms
- Gemini and Gemma text models for workflow prompts, with a default model that a workflow can override
- Refresh models loads separate transcription and text catalogs for your API key
Settings in this edition
- Enter your Gemini API key under API key, choose Save settings, and verify it with Check connection
- Choose Refresh models to load the transcription and text models available to your key
- Choose a Transcription model and set Transcription mode to Smart or Verbatim
- Choose the default Text model and set Temperature to Provider default or Custom (0 to 2)
Setup
- In TypeWhisper, open Settings > Discover plugins and install Google Gemini. The add-on then appears in the Settings sidebar, and its page opens.
- Enter your Gemini API key from Google AI Studio under API key, choose Save settings, and verify the key with Check connection.
- Choose Refresh models to load the transcription and text models available to your key. Until you refresh, the add-on offers Gemini 3.5 Transcribe and the Gemini Flash Lite, Flash and Pro “latest” text models.
- Choose a Transcription model. Under Transcription mode, choose Smart for cleaned-up text or Verbatim to keep the spoken wording. Live text while recording needs a transcription model that has a live version, such as Gemini 3.5 Transcribe.
- For text processing, choose the default Text model, set Temperature to Provider default or Custom, and if needed enter a value from 0 to 2 under Custom temperature. Choose Save settings.
- Under Settings > Dictation, select Google Gemini as the Provider; a Dictation Only workflow can also use its own transcription model. For text processing, select Google Gemini as a workflow’s Provider under Advanced or as the Default LLM; the workflow can choose a different Gemini or Gemma model.
Models and privacy
Transcription and text models are filtered into separate catalogs. Which models you can use depends on Google’s catalog and your API access. Opening the add-on’s page sends no request; only Refresh models fetches and saves the catalogs. The transcription provider does not support translation requests; translation can run as a separate text workflow.
Audio and text are sent to Google for the capability you select. Recorded audio is uploaded through Google’s Files API and transcribed with a request that sets store: false; the uploaded file is deleted on a best-effort basis after the request succeeds, fails, or is canceled. Billing and usage limits depend on your Google account.