Skip to content

Groq Whisper Speech to Text

Groq Whisper Speech to Text turns recorded audio files into text with the Whisper model, run on Groq’s servers, which makes it very fast. It recognises more than 99 languages and, if you don’t set the language, it detects it. It returns one Data per file, with the transcript in the text field.

You need a Groq API credential with your Groq account key. See Credentials. The audio is sent to Groq 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 Groq Whisper STT node. This component waits for the whole file.

When to use it

  • Turn voice notes into text when the answer has to come right away.
  • Transcribe audio in any language without knowing in advance which one.
  • 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

FieldWhat it does
API Key (required)Your Groq key. The key is filled in for you when you choose your Groq credential.
Audio (required)The audio (or video) file to transcribe: FLAC, MP3, MP4, M4A, OGG, WAV or WEBM. Connect it to the component that brings the file, for example a webhook, a chat with attachments or storage. It accepts several files.
LanguageOptional. The language of the audio, as its code, for example es or en. Setting it improves accuracy. Empty by default: Whisper detects the language.

Advanced options

FieldWhat it does
ModelThe Whisper model. Default: whisper-large-v3-turbo, the fastest and cheapest. whisper-large-v3 is the most accurate.

Output

The Data output returns one record per file:

{
"text": "Hola, buenos días. Su cita está confirmada para el martes a las 10 y media.",
"language": "es",
"file_name": "voice_note.ogg",
"provider": "Groq",
"model": "whisper-large-v3-turbo",
"duration": 5.1
}

duration is the length of the audio in seconds.

Each file can be up to 25 MB, about 25 minutes of MP3. If one is larger, the run stops with a message giving its size; compress it to MP3 or FLAC first.

Example

An estate agency receives WhatsApp voice messages from customers interested in a flat. The flow passes each audio to Groq Whisper Speech to Text, an agent reads the text, picks out the area and the budget, and replies at once with three flats that fit.

Tips

  • Whisper writes numbers and times as digits: “a las diez y media” comes out as “a las 10 y media” or “a las 10:30”.
  • 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 0.2 s, word for word.