Skip to content
Nappai Documentation

AssemblyAI

The AssemblyAI component lets you turn spoken words in an audio file into written text.
Just pick the file you want to transcribe, choose the language, and the component will send the file to AssemblyAI’s cloud service and return the spoken words as plain text. It’s useful for adding subtitles, creating searchable transcripts, or feeding the text into other parts of your workflow.

How it Works

When you run the component, it takes the audio file you selected and sends it to AssemblyAI’s API.
AssemblyAI processes the audio, converts the speech into text, and sends the result back.
The component then outputs that text as a Data object that can be used by other components in your dashboard.
Because the heavy lifting happens in AssemblyAI’s cloud, the component itself stays lightweight and fast.

Inputs

Before you can use the component, you must set up a credential:

  1. Go to Credentials in Nappai and add a new AssemblyAI API credential.
  2. Enter your AssemblyAI API key (you can get one from https://www.assemblyai.com/).
  3. In the component, choose that credential from the Credential dropdown.

Once the credential is selected, you can configure the following fields:

Input Fields

  • Audio File: The audio file you want to transcribe. Supported formats are mp3, mp4, wav, m4a, and mkv.
  • Message: A message that may contain files. This field can be used to pass additional data or context to the component.
  • Language: The language spoken in the audio file. Pick the code that matches the language (e.g., en for English, es for Spanish, fr for French, etc.). The default is English (en).

Outputs

  • Transcription Text: The spoken words from the audio file, returned as a Data object.
  • LLM Audio Transcriber: A reusable transcriber object (type LLMAudioVideoEncoder) that can be passed to other components that need to transcribe more audio later in the workflow.

Usage Example

  1. Add the component to your workflow.
  2. Select the AssemblyAI credential you created earlier.
  3. Upload an audio file (e.g., meeting.mp3).
  4. Choose the language (e.g., en).
  5. Run the workflow.
  6. The Transcription Text output will contain the full transcript, which you can then feed into a text‑analysis component or display in a dashboard panel.
  • Text Summarizer – Take the transcript and create a concise summary.
  • Keyword Extractor – Pull out important terms from the transcript for tagging or search.
  • Audio Player – Play the original audio file back to the user.

Tips and Best Practices

  • Keep audio files under the size limit recommended by AssemblyAI (usually 25 MB for free accounts).
  • Use the correct language code to improve transcription accuracy.
  • If you have multiple audio files, consider using the Mapping Mode (if available) to feed them in a loop.
  • Store the transcript in a database or a file for future reference.

Security Considerations

  • The AssemblyAI API key is stored securely in Nappai’s credential store.
  • Never expose the API key in the workflow or in any public-facing component.
  • Use role‑based access control to restrict who can create or edit the credential.