EchoInStone/main.py at main · jeanjerome/EchoInStone · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
import argparse
from EchoInStone.utils import configure_logging
import logging

from EchoInStone.capture.downloader_factory import get_downloader
from EchoInStone.processing import AudioProcessingOrchestrator, WhisperAudioTranscriber, PyannoteDiarizer, SpeakerAligner
from EchoInStone.utils import DataSaver
from EchoInStone.utils import timer, log_time

# Configure logging
configure_logging(logging.INFO)
logger = logging.getLogger(__name__)


# Youtube = 'https://www.youtube.com/watch?v=ipXG9iQq-Tw'
# Podcast = 'https://radiofrance-podcast.net/podcast09/rss_13957.xml'

# Test Youtube video = 'https://www.youtube.com/watch?v=plZRCMx_Jd8'
# Test Podcast = 'https://radiofrance-podcast.net/podcast09/rss_13957.xml'
# Test MP3 File = 'https://media.radiofrance-podcast.net/podcast09/25425-13.02.2025-ITEMA_24028677-2025C53905E0006-NET_MFC_D378B90D-D570-44E9-AB5A-F0CC63B05A14-21.mp3'


@timer
def main(echo_input, output_dir, transcription_output):
    """
    Main function to orchestrate the audio processing pipeline.
    """
    # Initialize components
    downloader = get_downloader(echo_input, output_dir)
    transcriber = WhisperAudioTranscriber()
    diarizer = PyannoteDiarizer()
    aligner = SpeakerAligner()
    data_saver = DataSaver(output_dir=output_dir)

    # Create an instance of AudioProcessingOrchestrator
    orchestrator = AudioProcessingOrchestrator(downloader, transcriber, diarizer, aligner, data_saver)

    # Process the input URL
    logger.info("Starting transcription process...")
    speaker_transcriptions = orchestrator.extract_and_transcribe(echo_input)
    if speaker_transcriptions:
        # Save the results to a file
        data_saver.save_data(transcription_output, speaker_transcriptions)

        logger.info(f"Transcriptions have been saved to {transcription_output}")

        # Display the results
        for speaker, start_time, end_time, segment_text in speaker_transcriptions:
            logger.info(f"Speaker {speaker} ({start_time:.2f}s to {end_time:.2f}s): {segment_text}")
    else:
        logger.warning("No transcriptions were generated.")

if __name__ == "__main__":
    parser = argparse.ArgumentParser(description="EchoInStone Audio Processing CLI")
    parser.add_argument("echo_input", type=str, help="URL of the audio input (YouTube, podcast, or direct audio file)")
    parser.add_argument("--output_dir", type=str, default="results", help="Directory to save the output files")
    parser.add_argument("--transcription_output", type=str, default="speaker_transcriptions.json", help="Filename for the transcription output")

    args = parser.parse_args()

    main(args.echo_input, args.output_dir, args.transcription_output)