AudioProcessor 0.1.4

A Python package for recording, transcribing, and converting audio

AudioProcessor

AudioProcessor is a Python class that provides functionality for recording audio, transcribing audio to text, and converting text to audio.

Class Methods

__init__(self, frame_duration_ms=30, sample_rate=16000, chunk_size=1024, vad_mode=2)

Initialize the AudioProcessor class.

Parameters:

• frame_duration_ms: Frame duration in milliseconds (default: 30 ms)
• sample_rate: Audio sample rate (default: 16000 Hz)
• chunk_size: Audio chunk size for processing (default: 1024)
• vad_mode: Voice Activity Detection (VAD) mode (default: 2)

record_audio(self, output_file_path)

Record audio from the microphone and save it to a .wav file.

Parameters:

• output_file_path: Output file path for the recorded audio

audio_to_text(self, audio_file_path, **kwargs)

Transcribe an audio file.

Parameters:

• audio_file_path: Input audio file path
• **kwargs: Keyword arguments for preprocessing and other options

Returns:

• Transcribed text as a string

text_to_audio(self, text, audio_file_path, lang='en', volume=1.0, sample_rate=44100, bit_depth=16)

Convert text to an audio file and play it.

Parameters:

• text: Input text to convert to audio
• audio_file_path: Output audio file path
• lang: Language of the text (default: 'en')
• volume: Volume level of the output audio (default: 1.0)
• sample_rate: Audio sample rate (default: 44100 Hz)
• bit_depth: Audio bit depth (default: 16)

Private Methods

These methods are used internally by the class and should not be called directly by the user.

_save_audio_to_file(self, audio, file_path)

Save the audio data to a .wav file.

_process_audio_text(self, audio_text, **kwargs)

Process the audio_text and return the transcribed text.

_parse_recognition_response(self, response, show_all)

Parse the response from the speech recognizer.

_process_recognized_audio(self, audio)

Process the recognized audio and return the transcribed text.

_play_audio_file(self, audio_file_path, volume=1.0, normalize=False)

Play the audio file with the specified volume and optional normalization.

_generate_frames(self, audio, sample_rate, frame_duration_ms)

Generate audio frames from raw audio data.

_collect_voiced_segments(self, frames, vad, sample_rate, ratio=2)

Collect voiced segments from audio frames using the Voice Activity Detection (VAD) algorithm.