pyaudiocook 1.0.1

A Python package for audio recording and transcribing

pyaudiocook

A Python package for recording and transcribing audio.

• pyaudiocook
  • Installation
  • Usage
    • AudioRecorder
      • Example: start recording
      • Example: pause/resume recording
      • Example: stop recording
    • AudioTranscriber
      • Example: Local whisper model and single audio file
      • Example: Local whisper model and multiple audio file
      • Example: Calling OpenAI API
      • Example: Using another model

Installation

pip install pyaudiocook

Usage

The package contains two classes: AudioRecorder and AudioTranscriber.

AudioRecorder

Example: start recording

recorder = AudioRecorder()
recorder.start_recording()

Example: pause/resume recording

recorder = AudioRecorder()
recorder.toggle_recording()

Example: stop recording

recorder = AudioRecorder()
recorder.stop_recording()

AudioTranscriber

Before getting started, make sure you have already obtained an API key on OpenAI. If you have not done so, follow this guide to create one.

Example: Local whisper model and single audio file

# Transcribe using local whisper model, which is the default
transcriber = AudioTranscriber('audio_1.wav')
transcriber.transcribe()
print(transcriber.texts)

Example: Local whisper model and multiple audio file

# You can pass multiple audio files for transcribing
transcriber = AudioTranscriber('audio_1.wav', 'audio_2.wav')
transcriber.transcribe()
print(transcriber.texts)

Example: Calling OpenAI API

Note: API usage fee applies.

# You can pass multiple audio files for transcribing
transcriber = AudioTranscriber('audio_1.wav', 'audio_2.wav', mode='remote')
transcriber.transcribe()
print(transcriber.texts)

Example: Using another model

Note: this only applies when local whisper model is used

# Default is tiny.en
transcriber = AudioTranscriber('audio_1.wav', 'audio_2.wav', mode='local', model='base.en')
transcriber.transcribe()
print(transcriber.texts)