This guide demonstrates how to transcribe a local audio file using the AssemblyAI API. You’ll learn how to set up your environment, configure your API key, and transcribe an audio file stored on your local machine.
assemblyai-local-transcription.py
Copy
import osfrom dotenv import load_dotenvimport assemblyai as aai# Load environment variables from .env fileload_dotenv()# Get AssemblyAI API key from environment variablesaai.settings.api_key = os.getenv('ASSEMBLYAI_API_KEY')# Initialize the transcribertranscriber = aai.Transcriber()# Define the local audio file pathlocal_audio_file_path = "downloads/voice.m4a" # Ensure this path is correct and the file exists# Set the transcription configurationconfig = aai.TranscriptionConfig(speaker_labels=True)try: # Transcribe the local audio file transcript = transcriber.transcribe(local_audio_file_path, config) # Print the transcription text and ID print(transcript.text) print(transcript.id) # Print the speaker labels and their corresponding text for utterance in transcript.utterances: print(f"Speaker {utterance.speaker}: {utterance.text}")except FileNotFoundError as e: print(f"File not found: {e.filename}")except Exception as e: print(f"An error occurred: {str(e)}")
Create a Python script named assemblyai-local-transcription.py with the following content:
assemblyai-local-transcription.py
Copy
import osfrom dotenv import load_dotenvimport assemblyai as aai# Load environment variables from .env fileload_dotenv()# Get AssemblyAI API key from environment variablesaai.settings.api_key = os.getenv('ASSEMBLYAI_API_KEY')# Initialize the transcribertranscriber = aai.Transcriber()# Define the local audio file pathlocal_audio_file_path = "downloads/voice.m4a" # Ensure this path is correct and the file exists# Set the transcription configurationconfig = aai.TranscriptionConfig(speaker_labels=True)try: # Transcribe the local audio file transcript = transcriber.transcribe(local_audio_file_path, config) # Print the transcription text and ID print(transcript.text) print(transcript.id) # Print the speaker labels and their corresponding text for utterance in transcript.utterances: print(f"Speaker {utterance.speaker}: {utterance.text}")except FileNotFoundError as e: print(f"File not found: {e.filename}")except Exception as e: print(f"An error occurred: {str(e)}")
You have successfully transcribed a local audio file using the AssemblyAI API! This guide provided a basic example to get you started. You can now expand on this by customizing the transcription configuration and handling different audio files stored locally.