# Build voice-first apps with Gemini 3.5 Transcribe

## Executive summary

Google launched Gemini 3.5 Transcribe, an advanced LLM-based model designed for building voice-first applications. This model is available via both the Interactions API and the Live API, offering fast, contextually accurate transcription of multi-speaker recordings. Key strengths include superior recognition of structured data like email addresses and phone numbers, as well as robust support for over 70 different languages.

## Key takeaways

- Model Availability: Gemini 3.5 Transcribe is available on both the Interactions API and the Live API.
- Structured Data Recognition: The LLM-based model excels at transcribing alphanumerics, such as email addresses (e.g., thorwebdev@google.com) and recognizing correct US phone number formats.
- Multi-Language Support: The model can recognize and transcribe over 70 different languages, even when language hints are set to English.

## Technical details

- Model Architecture & Function: Gemini 3.5 Transcribe is an LLM-based transcription model that provides deep audio understanding and reasoning capabilities.
- API Integration: The service can be integrated into applications using the Interactions API or for real-time use via the Live API.

## Practical implications

- Build voice-first interfaces that require highly accurate, contextually rich transcription.
- Implement robust data extraction from audio recordings (e.g., automatically capturing emails or phone numbers).
- Develop multi-lingual applications requiring reliable speech recognition across diverse language sets.

## Topics

Voice AI, Gemini 3.5 Transcribe, Speech Recognition, Interactions API, Live API, Google Developers

Source: https://www.youtube.com/watch?v=qzLrKKjdsPU