How to Use AssemblyAI

Getting started with AssemblyAI

In this guide, you will learn how to use AssemblyAI to transcribe audio files and detect speakers. By the end, you’ll be able to set up your account, make your first API call, and improve your transcription results.

Step 1: Sign up and set up

1. Go to AssemblyAI's website.
2. Click on the Get Started for Free button.
3. Fill in the required information to create your account.
4. Once you are logged in, navigate to the API Keys section in the dashboard to find your unique API key.

Step 2: Your first transcription

1. Prepare an audio file (WAV, MP3, etc.) that you want to transcribe.
2. Use a tool like Postman or cURL to make your API call.
3. Construct a POST request to https://api.assemblyai.com/v2/transcript with the following JSON body:

{
     "audio_url": "YOUR_AUDIO_FILE_URL"
   }

1. Include your API key in the headers:

Authorization: YOUR_API_KEY

1. Send the request. You’ll receive a JSON response containing a transcript_id. Use this ID to check the status of your transcription by sending a GET request to https://api.assemblyai.com/v2/transcript/YOUR_TRANSCRIPT_ID.

Step 3: Get better results

1. To enhance your transcription, include optional parameters in your POST request:

- For speaker detection, add "speaker_labels": true to your JSON body. - For automatic punctuation, add "punctuate": true. - You can also specify language with "language_code": "en-US" for English.

1. Check the status of your transcription by sending a GET request periodically until you see "status": "completed" in the response.

Pro tip

When testing, use short audio clips to speed up your workflow. This way, you can quickly iterate on your requests and adjust parameters without waiting long for results.

Common mistake to avoid

Ensure your audio file URL is publicly accessible. If it’s stored on a private server or requires authentication, AssemblyAI won’t be able to access it, resulting in errors.