askbob.speech.listener package¶

Submodules¶

askbob.speech.listener.file module¶

class askbob.speech.listener.file.FileUtteranceService(filename: str, aggressiveness: int = 1, lowpass_frequency: int = 65, highpass_frequency: int = 4000)¶

Bases: askbob.speech.listener.listener.UtteranceService

chunk: int = 320¶

askbob.speech.listener.listener module¶

class askbob.speech.listener.listener.UtteranceService(aggressiveness: int = 1, lowpass_frequency: int = 65, highpass_frequency: int = 4000)¶

Bases: object

The UtteranceService is responsible for compiling complete utterances for the speech transcriber.

block_size: int = 320¶

blocks_per_second: int = 50¶

buffer_queue: queue.Queue¶

channels: int = 1¶

property frame_duration_ms¶

input_rate: int¶

reset_buffer_queue: bool = True¶

sample_rate: int = 16000¶

utterances(padding_ms=300, ratio=0.75)¶

This is a generator that yields series of consecutive audio frames for each utterence, separated by a single None.

It determines the level of voice activity using the ratio of frames in padding_ms. It uses a buffer to include padding_ms prior to being triggered.

Example: (frame, …, frame, None, frame, …, frame, None, …)

utterence | | utterence |

vad: webrtcvad.Vad¶

write_wav(filename: str, data)¶

Writes audio frames to a .wav file.

Parameters

filename (str) – The filename of the .wav file to be written.
data – The audio frames.

askbob.speech.listener.mic module¶

class askbob.speech.listener.mic.MicUtteranceService(aggressiveness: int = 1, device_index: Optional[int] = None, input_rate: int = 16000, lowpass_frequency: int = 65, highpass_frequency: int = 4000)¶