You’ve probably used Gemini to analyze hours of video, summarize podcasts, or answer questions from recorded meetings (if you didn't you should, it's extremely useful!). But when all you need is a clean, hyper-accurate, and structured transcript from audio, spinning up a huge reasoning model with complicated prompts often feels like using a sledgehammer to crack a nut.
It's Google's dedicated speech-to-text model built on Gemini's audio understanding core, optimized specifically for fast, accurate, and cost-effective transcription. Whether you want an exact court-reporter transcript with millisecond timestamps, or a reading-optimized summary that removes all your awkward "ums" and "uhs", this model handles it natively with zero prompt gymnastics.
🚀 Hands-on first: If you want to jump straight into running the code yourself, open the interactive Gemini Transcribe Colab notebook! It's ready to run so you can dirrectly experience how the model work. Prefer a visual UI with zero coding? You can also test speech recognition directly in Google AI Studio.
Here's what you'll find in this guide: Why a Dedicated Transcription Model? (Audio Understanding vs. Transcribe) Setup & The Files API Steer Languages & Code-Switching (85+ Locales) Custom Vocabulary: Never Misspell Technical Jargon Again The Killer Feature: Smart Transcription vs. Verbatim Mode Speaker Diarization: Who Said What? Word-Level Timestamps: Precise Time Offsets for Every Spoken Word Decision Matrix: Which Configuration Should You Use? What About Real-Time Live Streaming? Why a Dedicated Transcription Model?
Before looking at the code, let's get the mental model straight. You might wonder: "Can't I just upload an MP3 to Gemini 3.7 and say 'Transcribe this'?"
| Feature | General Audio Understanding (e.g. Gemini 3.7) | Dedicated Transcribe (gemini-3.5-transcribe) | | :--- | :--- | :--- | | Primary Job | Reasoning, Q&A, sentiment analysis, audio chat | High-throughput, precise speech-to-text | | Speaker Diarization | Prompt-dependent (can hallucinate turns) | Native segment labeling (spk:0, spk:1) | | Timestamps | Approximate timecodes via text prompt | True word-level millisecond offsets in metadata | | Vocabulary Biasing | System prompt instructions | Native acoustic biasing dictionary (up to 1,000 terms) | | Cost & Latency | Full multimodal LLM generation overhead | Optimized lightweight speech pipeline |
Pro tip: If you need to ask questions about what happened in an audio file ("What was the action item for Alice?"), use a multimodal model like Gemini 3.7. If you need the transcript itself, subtitles, or cleaned dictation notes, use Gemini Transcribe! Setup & The Files API
The Gemini 3.5 Transcribe model runs on the modern Google GenAI SDK (google-genai v2.0+) using the Interactions API.
Make sure you have an API key from Google AI Studio, set it as GEMINIAPIKEY, and let's look at how audio gets passed to the model:
Watch the demo video below to see the baseline transcription in action—handling natural speech and bilingual code-switching with ease:
Why use the Files API? When dealing with audio and video, you never want to inline raw audio bytes as base64 in your API requests—it blows up the payload size by 33%, easily hits network timeouts, and requires re-uploading the same bytes if you want to rerun a query.
The Files API solves this cleanly: Large file support: Upload audio and video files up to 2 GB per file (with 20 GB of total project storage). Temporary lifecycle: Files are stored for 48 hours and automatically cleaned up afterwards. It's completely free! Storage and uploads in the Files API incur zero additional cost—you only pay for token processing when you actually run inference against the model. Steer Languages & Code-Switching (85+ Locales)
As you saw in the video above, Gemini Transcribe automatically identifies spoken languages out of the box and seamlessly handles code-switching (when someone mixes multiple languages in the same sentence—like switching between French and English mid-sentence, which happens to me all the time!).
However, if you know your audio is exclusively in a specific language or regional dialect, you can pass explicit BCP-47 language codes in transcriptionconfig to bias recognition:
Note: Leaving languagecodes=[] (or omitting it) enables full automatic detection across 85+ supported languages and locales. Check out the Audio Transcription Documentation for the complete list of language codes. Custom Vocabulary: Never Misspell Technical Jargon Again
Every developer has suffered from an ASR model mangling proper names, confusing specialized libraries with everyday dictionary words (turning "ScaNN" into "scan", or "Qdrant" into "quadrant"), or inventing phonetically similar terms ("Sitsi" instead of "CitC", "Thiago" instead of "Tiago").
With customvocabulary, you can pass a list of up to 1,000 domain-specific terms that the model will bias towards:
Watch the side-by-side comparison video below to see how the model behaves with and without custom vocabulary biasing: