import gradio as gr
from pyannote.audio import Pipeline

# טוען את המודל של Pyannote לזיהוי דוברים
pipeline = Pipeline.from_pretrained("pyannote/speaker-diarization")

# פונקציה לביצוע זיהוי דוברים
def diarize(audio_file):
    diarization = pipeline(audio_file)
    segments = []
    for turn, _, speaker in diarization.itertracks(yield_label=True):
        segments.append({
            "start": turn.start,
            "end": turn.end,
            "speaker": speaker
        })
    return segments

# הגדרת ממשק Gradio
interface = gr.Interface(
    fn=diarize, 
    inputs=gr.Audio(source="upload", type="filepath"), 
    outputs="json",
    description="Upload an audio file to get speaker diarization (timestamps and speaker IDs only)."
)

# הרצת הממשק
interface.launch()