gistfile1.txt

import collections
import sys
import wave
import time
import webrtcvad
import pyaudio

# Audio settings
FORMAT = pyaudio.paInt16
CHANNELS = 1
RATE = 16000
FRAME_DURATION = 30  # ms (10, 20, 30 allowed)
FRAME_SIZE = int(RATE * FRAME_DURATION / 1000)
PADDING_DURATION = 300  # ms
VAD_MODE = 2  # 0-3 (3 aggressive)

vad = webrtcvad.Vad(VAD_MODE)
audio = pyaudio.PyAudio()

stream = audio.open(format=FORMAT,
                    channels=CHANNELS,
                    rate=RATE,
                    input=True,
                    frames_per_buffer=FRAME_SIZE)

ring_buffer = collections.deque(maxlen=int(PADDING_DURATION / FRAME_DURATION))
triggered = False
voiced_frames = []

print("🎙 Listening...")

def save_wave(frames):
    filename = f"speech_{int(time.time())}.wav"
    wf = wave.open(filename, 'wb')
    wf.setnchannels(CHANNELS)
    wf.setsampwidth(audio.get_sample_size(FORMAT))
    wf.setframerate(RATE)
    wf.writeframes(b''.join(frames))
    wf.close()
    print(f"💾 Saved: {filename}")

try:
    while True:
        frame = stream.read(FRAME_SIZE, exception_on_overflow=False)
        is_speech = vad.is_speech(frame, RATE)

        if not triggered:
            ring_buffer.append((frame, is_speech))
            num_voiced = len([f for f, speech in ring_buffer if speech])

            if num_voiced > 0.8 * ring_buffer.maxlen:
                triggered = True
                print("🗣 Speech started")
                voiced_frames.extend([f for f, s in ring_buffer])
                ring_buffer.clear()
        else:
            voiced_frames.append(frame)
            ring_buffer.append((frame, is_speech))

            num_unvoiced = len([f for f, speech in ring_buffer if not speech])
            if num_unvoiced > 0.8 * ring_buffer.maxlen:
                print("🔇 Speech ended")
                save_wave(voiced_frames)
                triggered = False
                ring_buffer.clear()
                voiced_frames = []

except KeyboardInterrupt:
    print("Stopping...")

stream.stop_stream()
stream.close()
audio.terminate()