first pass at turn based conversation

vocode/turn_based/input_device/microphone_input.py

@@ -0,0 +1,59 @@
+from typing import Optional
+import sounddevice as sd
+import numpy as np
+from pydub import AudioSegment
+import io
+import wave
+
+from vocode.turn_based.input_device.base_input_device import BaseInputDevice
+
+
+class MicrophoneInput(BaseInputDevice):
+    DEFAULT_SAMPLING_RATE = 44100
+    DEFAULT_CHUNK_SIZE = 2048
+
+    def __init__(
+        self,
+        device_info: dict,
+        sampling_rate: int = None,
+        chunk_size: int = DEFAULT_CHUNK_SIZE,
+    ):
+        self.device_info = device_info
+        self.sampling_rate = sampling_rate or (
+            self.device_info.get("default_samplerate", self.DEFAULT_SAMPLING_RATE)
+        )
+        self.chunk_size = chunk_size
+        self.buffer: Optional[io.BytesIO] = None
+        self.wave_writer: Optional[wave.Wave_write] = None
+
+    def create_stream(self):
+        return sd.InputStream(
+            dtype=np.int16,
+            channels=1,
+            samplerate=self.sampling_rate,
+            blocksize=self.chunk_size,
+            device=int(self.device_info["index"]),
+            callback=self._stream_callback,
+        )
+
+    def _stream_callback(self, in_data: np.ndarray[np.int16], *_args):
+        audio_bytes = in_data.tobytes()
+        self.wave_writer.writeframes(audio_bytes)
+
+    def create_buffer(self):
+        in_memory_wav = io.BytesIO()
+        wave_writer = wave.open(in_memory_wav, "wb")
+        wave_writer.setnchannels(1)
+        wave_writer.setsampwidth(2)
+        wave_writer.setframerate(self.sampling_rate)
+        return in_memory_wav, wave_writer
+
+    def start_listening(self):
+        self.buffer, self.wave_writer = self.create_buffer()
+        self.stream = self.create_stream()
+        self.stream.start()
+
+    def end_listening(self) -> AudioSegment:
+        self.stream.stop()
+        self.buffer.seek(0)
+        return AudioSegment.from_wav(self.buffer)