Our Speech-to-Text model enables you to transcribe pre-recorded audio into written text.

On top of the transcription, you can enable other features and models, such as Speaker Diarization, by adding additional parameters to the same transcription request.

Quickstart

The following example transcribes an audio file.

1 import assemblyai as aai
2 
3 aai.settings.base_url = "https://api.assemblyai.com"
4 aai.settings.api_key = "YOUR_API_KEY"
5 
6 # Use a publicly-accessible URL
7 audio_file = "https://assembly.ai/wildfires.mp3"
8 
9 # Or use a local file:
10 # audio_file = "./example.mp3"
11 
12 config = aai.TranscriptionConfig(
13     speech_models=["universal-3-pro", "universal-2"],
14     language_detection=True,
15     speaker_labels=True,
16 )
17 
18 transcript = aai.Transcriber().transcribe(audio_file, config=config)
19 
20 if transcript.status == aai.TranscriptStatus.error:
21     raise RuntimeError(f"Transcription failed: {transcript.error}")
22 
23 print(f"\nFull Transcript:\n\n{transcript.text}")
24 
25 # Optionally print speaker diarization results
26 # for utterance in transcript.utterances:
27 #     print(f"Speaker {utterance.speaker}: {utterance.text}")

Example output

1 Smoke from hundreds of wildfires in Canada is triggering air quality alerts
2 throughout the US. Skylines from Maine to Maryland to Minnesota are gray and
3 smoggy. And...

API reference

You can find the API reference here.