server/AudioTranscriber.py

import io
from typing import Tuple
import threading
from multiprocessing import Queue
from queue import Empty
from faster_whisper import WhisperModel

import logging
import sys

# Configure logging
logging.basicConfig(
    level=logging.INFO,
    format="%(asctime)s - %(name)s - %(levelname)s - %(message)s",
    handlers=[logging.StreamHandler(sys.stdout)],
)

# Get a logger for your app
logger = logging.getLogger(__name__)


class AudioTranscriber(threading.Thread):
    def __init__(
        self,
        audio_queue: "Queue[Tuple[io.BytesIO, str]]",
        text_queue: "Queue[Tuple[str, str]]",
        language: str = "en",
        confidence_threshold: float = 0.5,
        device_index: int = 0,
    ):
        super().__init__()
        self.audio_queue = audio_queue
        self.action_queue = text_queue
        self.daemon = True  # Thread will exit when main program exits
        self.language = language
        self.confidence_threshold = confidence_threshold
        self.transcriber = WhisperModel(
            "large",
            device="cuda",
            device_index=device_index,
            compute_type="int8",
        )

    def run(self):
        while True:
            try:
                # Wait for 1 second before timing out and checking again
                audio_data, session_id = self.audio_queue.get(timeout=1)

                segments, _ = self.transcriber.transcribe(
                    audio_data, language=self.language
                )

                # Put the transcription results in the output queue
                for segment in segments:
                    if segment.no_speech_prob <= self.confidence_threshold:
                        self.action_queue.put((segment.text, session_id))
                        # Still print for debugging
                        logger.info(
                            f"[Thread {threading.get_ident()}] [{segment.start:.2f}s -> {segment.end:.2f}s] {segment.text}"
                        )
                    else:
                        self.action_queue.put(("", session_id))

            except Empty:
                continue  # If queue is empty, continue waiting
            except Exception as e:
                logger.error(f"Error processing audio chunk: {e}")