47 lines
1.4 KiB
Python
47 lines
1.4 KiB
Python
|
#!/usr/bin/env python3
|
||
|
|
"""
|
||
|
|
Speaker diarization using pyannote.audio.
|
||
|
|
"""
|
||
|
|
|
||
|
|
import json
|
||
|
|
import sys
|
||
|
|
from pathlib import Path
|
||
|
|
|
||
|
|
# Add backend to path
|
||
|
|
sys.path.insert(0, str(Path(__file__).parent))
|
||
|
|
|
||
|
|
from services.diarization import diarize_and_label
|
||
|
|
|
||
|
|
|
||
|
|
def main():
|
||
|
|
if len(sys.argv) < 2:
|
||
|
|
print("Usage: python diarization.py <command> [args...]", file=sys.stderr)
|
||
|
|
sys.exit(1)
|
||
|
|
|
||
|
|
command = sys.argv[1]
|
||
|
|
|
||
|
|
try:
|
||
|
|
if command == "diarize_and_label":
|
||
|
|
if len(sys.argv) < 4:
|
||
|
|
print("Usage: python diarization.py diarize_and_label <transcription_result_json> <audio_path> [hf_token] [num_speakers] [use_gpu]", file=sys.stderr)
|
||
|
|
sys.exit(1)
|
||
|
|
transcription_result = json.loads(sys.argv[2])
|
||
|
|
audio_path = sys.argv[3]
|
||
|
|
hf_token = sys.argv[4] if len(sys.argv) > 4 else None
|
||
|
|
num_speakers = int(sys.argv[5]) if len(sys.argv) > 5 and sys.argv[5] != "null" else None
|
||
|
|
use_gpu = sys.argv[6].lower() == "true" if len(sys.argv) > 6 else True
|
||
|
|
|
||
|
|
result = diarize_and_label(transcription_result, audio_path, hf_token, num_speakers, use_gpu)
|
||
|
|
print(json.dumps(result))
|
||
|
|
|
||
|
|
else:
|
||
|
|
print(f"Unknown command: {command}", file=sys.stderr)
|
||
|
|
sys.exit(1)
|
||
|
|
|
||
|
|
except Exception as e:
|
||
|
|
print(json.dumps({"error": str(e)}), file=sys.stderr)
|
||
|
|
sys.exit(1)
|
||
|
|
|
||
|
|
|
||
|
|
if __name__ == "__main__":
|
||
|
|
main()
|