Microphone

Microphone

Microphone Source for live audio recording of a connected microphone

__init__(self, device, format=8, channels=2, sampling_rate=44100, chunk_size=1024) special

Initialize the Source

Parameters:

Name	Type	Description	Default
device	str	Device id of the microphone	required
format		PyAudio format specification	8
channels	int	Number of channels of the device	2
sampling_rate	int	The audio sampling rate	44100
chunk_size	int	Size of the chunks of the recordings	1024

Source code in multisensor_pipeline/modules/audio/microphone.py

def __init__(self, device: str, format=pyaudio.paInt16, channels: int = 2, sampling_rate: int = 44100,
             chunk_size: int = 1024):
    """
    Initialize the Source
    Args:
       device: Device id of the microphone
       format: PyAudio format specification
       channels: Number of channels of the device
       sampling_rate: The audio sampling rate
       chunk_size: Size of the chunks of the recordings
    """
    super(Microphone, self).__init__()

    self.device = device
    self.format = format
    self.channels = channels
    self.sampling_rate = sampling_rate
    self.chunk_size = chunk_size

    self._mic = pyaudio.PyAudio()
    self._stream = self._mic.open(format=self.format,
                                  channels=self.channels,
                                  rate=self.sampling_rate,
                                  input=True,
                                  frames_per_buffer=self.chunk_size)

on_stop(self)

Stops the Microphone source and closes the stream

Source code in multisensor_pipeline/modules/audio/microphone.py

def on_stop(self):
    """
    Stops the Microphone source and closes the stream
    """
    self._stream.stop_stream()
    self._stream.close()
    self._mic.terminate()

on_update(self)

Sends chunks of the audio recording

Source code in multisensor_pipeline/modules/audio/microphone.py

def on_update(self) -> Optional[MSPDataFrame]:
    """
    Sends chunks of the audio recording
    """
    data = self._stream.read(self.chunk_size)
    return MSPDataFrame(topic=self._generate_topic(name="audio"), chunk=data)