Intelligent real-time transcription
The model is built to transcribe speech into polished text even when audio includes noise, jargon, or self-corrections.
Gemini 3.5 Transcribe
Gemini 3.5 Transcribe is Google’s speech-to-text model for real-time and recorded audio transcription, with multilingual support, timestamps, and speaker attribution.
Overview
Gemini 3.5 Transcribe is Google’s speech-to-text model for intelligent transcription. The announcement positions it as a more precise option for real-time voice interactions and for converting recorded audio into clean, formatted text.
The model is designed to handle common transcription problems such as background noise, jargon, disfluencies, and mid-sentence corrections. It is available through the Gemini API in Google AI Studio and the Gemini Enterprise Agent Platform, and it also powers voice features in Google products such as the Gemini app on macOS and Rambler on Android.
Key capabilities
Low-latency streaming via Live API
The Live API supports continuous bidirectional streaming with sub-second latency for interactive voice applications.
Pre-recorded audio processing
The Interactions API handles recorded audio, meetings, and call logs, and can return speaker attribution plus word-level timestamps.
Multilingual language detection
The model automatically detects and transcribes more than 85 languages, including regional accents and dialects, and can handle live language switches.
Custom vocabulary support
It recognizes specialized terms and unique spellings through custom vocabulary supplied by the user.
Voice-enabled function calling
The model can delegate tasks such as image generation and file analysis through function calls, and this is currently available in the Gemini macOS app.
Practical use cases
Interactive voice applications
Build voice agents and other interactive apps that need continuous speech-to-text with low latency and bidirectional streaming.
Meeting and call transcription
Transcribe meetings, call recordings, and audio logs, then use speaker attribution and timestamps to review what was said and by whom.
Live captioning
Create real-time captioning tools that need transcription output while audio is still being captured.
Voice-driven productivity
Use voice commands in the Gemini app on macOS or Android workflows to dictate, edit, summarize, or clean up speech into polished text.
Post-call analytics
Process recorded audio for post-call analytics pipelines where accurate formatted text is needed for downstream analysis.
Pros and Cons
Pros
- Supports both real-time streaming and pre-recorded transcription workflows.
- Includes speaker attribution and word-level timestamps for recorded audio.
- Automatically detects and transcribes more than 85 languages.
- Handles background noise, jargon, filler words, and self-corrections.
- Available to developers through Google AI Studio and the Gemini Enterprise Agent Platform.
Cons
- Pricing is not published on the source page.
- Some consumer features are only partially available, such as Rambler on Android in select countries and languages, Chrome coming soon, and 3+ speaker support marked experimental.
FAQ
How do developers access Gemini 3.5 Transcribe?
Developers can access Gemini 3.5 Transcribe in the Gemini API through Google AI Studio and the Gemini Enterprise Agent Platform. The announcement also says it is in public preview for developers.
What kinds of transcription workflows does it support?
The page describes two API paths: the Live API for real-time streaming with sub-second latency, and the Interactions API for pre-recorded audio with speaker attribution and word-level timestamps.
What transcription challenges is it designed to handle?
It is designed to handle background noise, complex jargon, disfluency cleanup, custom vocabulary, and more than 85 languages, including regional accents and dialects.
Where is Gemini 3.5 Transcribe available in Google products?
The announcement says the model is available to everyone in the Gemini app on macOS in English, in Rambler on Android in select countries and languages, and is coming soon to Chrome.
Is pricing published for Gemini 3.5 Transcribe?
The source does not provide public pricing details. It only states that the model is in public preview across developer and enterprise access points.
Quick Facts
- Category
- Speech-to-text / transcription
- Primary users
- Developers, enterprises, and consumer app users
- Access
- Gemini API in Google AI Studio; Gemini Enterprise Agent Platform; Gemini app on macOS; Rambler on Android
- API modes
- Live API for real-time streaming; Interactions API for pre-recorded audio
- Languages
- 85+ languages with regional accents and dialects
- Source domain
- blog.google
Gemini 3.5 Transcribe Alternatives
SpeakoFlow
SpeakoFlow is a free, open-source desktop voice-to-text app for Windows, macOS, and Linux. Dictate into any app, use Flow, cleanup, translation, and a screen-aware assistant.
QuickQuill
QuickQuill is a local-first macOS dictation and transcription app to record meetings, summarize audio, and export notes without the cloud.
Zen Whisper
Zen Whisper is a local-first dictation and transcription app for Apple Silicon Macs on macOS 14+. Dictate into most Mac text fields and transcribe media or supported public links across 110 languages.
Speech to Text Converter
Speech to Text Converter is a browser-based transcription tool for live dictation and uploaded audio or video files. Free for short tasks, Pro offers unlimited transcription, AI summaries, translation, speaker ID, and advanced exports.
Sanota
Sanota turns spoken memories and interviews into clear written stories for personal storytelling, family history and shared memories, with guided prompts and subscriptions.
Carbon Voice
Carbon Voice is an async voice messaging app for teams and individuals, with transcripts, AI catch-up, and cross-device access.