This tutorial will guide you through setting up Google Gemini's Live API for use with Sokuji. Gemini offers excellent multilingual support and automatic audio processing capabilities.
Visit Google AI Studio and sign in with your Google account.
In Google AI Studio:
The API key should automatically have access to Gemini models. If you encounter issues, visit the Google Cloud Console and ensure the Gemini API is enabled for your project.
Open Sokuji and navigate to the Settings panel:
gemini-2.0-flash-live-001
)Aoede
, Puck
, or Charon
)Start a translation session to verify everything is working:
Gemini handles turn detection automatically, so you don't need to configure threshold or silence duration settings. The model intelligently determines when you've finished speaking.
Gemini offers 30 unique voice personalities with different characteristics. Experiment with different voices to find one that suits your use case:
API Key Error: Ensure your API key is valid and has not expired. Check that the Gemini API is enabled in your Google Cloud project.
Quota Exceeded: Check your usage limits in Google AI Studio and consider upgrading if needed.
Language Not Supported: Verify that both your source and target languages are supported by Gemini. Some language pairs may have better quality than others.