Files
tueit_Transkriptor/audio.py

47 lines
1.3 KiB
Python

import wave
import threading
import numpy as np
class AudioRecorder:
def __init__(self, sample_rate: int = 16000):
self.sample_rate = sample_rate
self._buffer: list[np.ndarray] = []
self._stream = None
self.is_recording = False
self._lock = threading.Lock()
def _callback(self, indata, frames, time, status):
if self.is_recording:
with self._lock:
self._buffer.append(indata[:, 0].copy().astype(np.int16))
def start(self):
import sounddevice as sd
self._buffer = []
self.is_recording = True
self._stream = sd.InputStream(
samplerate=self.sample_rate,
channels=1,
dtype="int16",
callback=self._callback,
)
self._stream.start()
def stop(self):
self.is_recording = False
if self._stream:
self._stream.stop()
self._stream.close()
self._stream = None
def save_wav(self, path: str) -> str:
with self._lock:
data = np.concatenate(self._buffer) if self._buffer else np.zeros(0, dtype=np.int16)
with wave.open(path, "wb") as wf:
wf.setnchannels(1)
wf.setsampwidth(2)
wf.setframerate(self.sample_rate)
wf.writeframes(data.tobytes())
return path