AssemblyAI

AssemblyAI transcribes audio files into text using the AssemblyAI service. You give it one or more recordings, choose the language being spoken, and get the text of each one, ready to summarise, analyse or store.

You need an AssemblyAI account and its access key.

When to use it

Transcribe the voice notes customers send through a messaging channel.
Turn meeting recordings into text to produce minutes afterwards.
Convert customer service calls into text to analyse or classify them.
Prepare the text of interviews or podcasts to publish or summarise.

Configuration

Field	What it does
Assembly API Key (required)	Your AssemblyAI access key, available from your account on assemblyai.com.
Audio Input	The audio file or files to transcribe. Connect the file coming from another component, for example an uploaded file or a message attachment.
Language	Language spoken in the audio. Default: `en` (English). Options: `en`, `es`, `fr`, `de`, `it`, `pt`, `ja`, `ko`, `hi`, `ar`, `ru`, `zh`.

This component supports Batch Mode, to process a list of audio files one by one, and it can also be used as an agent tool. See Key concepts.

Output

The Data output returns one result per audio file with:

text: the transcription.
language: the language you selected.

If a file cannot be transcribed, its result includes an error field with the reason, and the other files are processed as usual.

Example

A clinic receives voice messages from patients who want to book an appointment.

Connect the incoming audio to Audio Input.
Choose es in Language if your patients speak Spanish.
Send the transcription to an agent that identifies the reason for the visit and suggests a free slot.

Good practices

Always choose the actual language of the audio: transcription quality depends on it.
If you receive audio in several languages, detect the language first or separate the files by channel.