Gemini Speech to Text
Gemini Speech to Text turns recorded audio files into text with a Google Gemini model. It transcribes word for word in the language spoken, without translating or summarising, and detects the language if you don’t set it. It returns one Data per file, with the transcript in the text field.
You need a Google Gemini credential with your Google AI key, the same one the Gemini model uses. See Credentials. The audio is sent to Google to be transcribed.
This is not the voice node for calls: to listen to a caller during a live call, the Voice Agent uses the Gemini STT node. This component waits for the whole file.
When to use it
- Transcribe long recordings, such as a meeting lasting several hours.
- Turn audio in any language into text when you already work with Gemini.
- Let an agent transcribe audio when it needs to, using the component as a tool.
- Transcribe a list of audio files in one go with Batch Mode.
Configuration
| Field | What it does |
|---|---|
| API Key (required) | Your Google AI key. The key is filled in for you when you choose your Google Gemini credential. |
| Audio (required) | The audio (or video) file to transcribe: WAV, MP3, OGG, M4A, WEBM… Connect it to the component that brings the file, for example a webhook, a chat with attachments or storage. It accepts several files. |
| Language | Optional. The expected language of the audio, for example es or en. Empty by default: Gemini detects the language. |
Advanced options
| Field | What it does |
|---|---|
| Model | The Gemini model. Default: gemini-3.8-flash, the most accurate. gemini-3.5-flash-lite is cheaper and faster. |
Output
The Data output returns one record per file:
{ "text": "Hola, buenos días. Su cita está confirmada para el martes a las 10:30.", "language": "es", "file_name": "meeting.mp3", "provider": "Gemini", "model": "gemini-3.8-flash"}The text comes out as one block: no speaker names and no timestamps.
Each file can be up to 2 GB, about 9 hours of audio. Files over 14 MB are first uploaded to Google and deleted as soon as the transcription ends.
Example
A law firm records its meetings with clients. When each one ends, the flow takes the recording from storage, passes it to Gemini Speech to Text, and an agent writes the minutes from the text and emails them to the attendees.
Related components
- Speech to text and text to speech: every component that transcribes audio or produces it.
- Gemini Text to Speech: the opposite, turns text into audio with the same credential.
- Deepgram Speech to Text and Groq Whisper Speech to Text: they transcribe faster.
Tips
- Gemini writes numbers and times as digits: “a las diez y media” comes out as “a las 10 y media” or “a las 10:30”.
- It is the slowest of the five new speech-to-text components. Measured in our test environment on 3 November 2026: the sentence “Hola, buenos días. Su cita está confirmada para el martes a las diez y media.” was transcribed in 4.5 s, word for word. If the answer has to be immediate, choose another one.