Get Transcription Result

import requests
import json
import pandas as pd
from datetime import timedelta

# Authentication details
headers = {
    "x-api-key": "your-api-key",  # Replace with your actual API key
    "Content-Type": "application/json"
}

def get_transcription(run_id, format_type="txt", data_type="file"):
    """
    Fetches the complete transcription for a processed media file.
    
    Parameters:
        run_id (int): The unique identifier for your transcription task
        format_type (str): Output format - 'txt', 'srt' or 'vtt'
        data_type (str): How to receive data - 'file' for URL or 'json' for raw data
    
    Returns:
        If data_type is 'file': A URL to download the transcription file
        If data_type is 'json': An array of transcription segments with timing and speaker information
    """
    try:
        # Build query parameters based on user preferences
        params = {
            "format_type": format_type,
            "data_type": data_type
        }
        
        # Request the transcription with specified format and type
        response = requests.get(
            f"https://client.camb.ai/apis/transcription-result/{run_id}",
            headers=headers,
            params=params  # Include our format preferences in the request
        )
        
        # Ensure successful response
        response.raise_for_status()
        
        # Parse the response based on the requested data_type
        result = response.json()
        
        if data_type == "file":
            # For file requests, we receive a URL to download the transcript
            print(f"Successfully retrieved transcription file URL in {format_type} format")
            print(f"Download URL: {result['file_url']}")
            print(f"URL will expire in 24 hours - download your file soon")
            return result['file_url']
        else:
            # For JSON requests, we receive the transcription data directly
            transcription_data = result['segments']
            print(f"Successfully retrieved transcription with {len(transcription_data)} segments")
            return transcription_data
        
    except requests.exceptions.RequestException as e:
        print(f"Error retrieving transcription: {e}")
        if hasattr(e, 'response') and e.response is not None:
            print(f"Response details: {e.response.text}")
        return None

# Example 1: Get a downloadable SRT file for subtitle integration
run_id = 12345
srt_url = get_transcription(run_id, format_type="srt", data_type="file")
print(f"You can now download the SRT file and import it directly into your video editor")

# Example 2: Get raw JSON data for analysis
run_id = 12345
transcription = get_transcription(run_id, format_type="txt", data_type="json")

# Let's analyze the JSON data if we received it
if transcription and isinstance(transcription, list):
    # Convert to DataFrame for easier manipulation
    df = pd.DataFrame(transcription)
    
    # Format timestamps as human-readable
    df['duration'] = df.apply(lambda row: round(row['end'] - row['start'], 2), axis=1)
    df['start_formatted'] = df['start'].apply(lambda x: str(timedelta(seconds=x)))
    df['end_formatted'] = df['end'].apply(lambda x: str(timedelta(seconds=x)))
    
    # Display a sample of the transcription
    print("\nTranscription Preview:")
    for i, row in df.head(5).iterrows():
        print(f"[{row['start_formatted']} → {row['end_formatted']}] {row['speaker']}: {row['text']}")
    
    # Generate some basic analytics
    print("\nTranscription Statistics:")
    print(f"Total duration: {str(timedelta(seconds=df['end'].max()))}")
    print(f"Unique speakers: {df['speaker'].nunique()}")
    print(f"Most frequent speaker: {df['speaker'].value_counts().idxmax()}")
    
    # Export to various formats
    df.to_csv("transcription.csv", index=False)
    print("\nTranscription exported to CSV for further analysis")