from typing import Optional
from faster_whisper import WhisperModel


_model: Optional[WhisperModel] = None


def get_model(model_size: str = "medium", device: str = "cpu", compute_type: str = "int8") -> WhisperModel:
    global _model
    if _model is None:
        _model = WhisperModel(model_size, device=device, compute_type=compute_type)
    return _model


def transcribe(audio_path: str, model_size: str = "medium", device: str = "cpu", compute_type: str = "int8", language: str = "ru") -> tuple[list[dict], str]:
    model = get_model(model_size, device, compute_type)

    segments, info = model.transcribe(audio_path, language=language, beam_size=5, vad_filter=True)

    result = []
    full_text_parts = []
    for seg in segments:
        result.append({
            "start": round(seg.start, 2),
            "end": round(seg.end, 2),
            "text": seg.text.strip(),
        })
        full_text_parts.append(seg.text.strip())

    full_text = " ".join(full_text_parts)
    return result, full_text