PyBreathTranscript 1.0.0

Tool for transcribing breaths from audio files

PyBreathTranscript

This package is used for transcribing breath recordings (captured using an in-ear microphone) into text symbols.

Check Full project to see this library in action.

Quickstart

Install the package:

pip install -i https://test.pypi.org/simple/ PyBreathTranscript

or

pip install git+https://github.com/ilya-shlom/PyBreathTranscript.git

Example usage:

from PyBreathTranscript import transcript as bt

result = bt.transcript("breath.wav", bt.FINGERPRINT)
print(result)

Usage

Import library into your project using

from PyBreathTranscript import transcript as bt

To transcribe a recording of your breath, use the bt.transcript(file_path, method) function:

transcript = bt.transcript("path/to/your/recording.wav", bt.FINGERPRINT)

file_path: str - path of your recording.

method: int - method used to transcribe an audio file. There are two methods available now acceptable by constants:

1. bt.FINGERPRINT – method based on an audio fingerprinting algorithm. This is the recommended and default method.

2. bt.WAVEFORM – compares waveforms and returns the best-matching symbol. This approach is more realistic but less distinctive, recommended for demonstration purposes only.

---

Please note: this library is created for educational and demonstration purposes. For the best experience, an in-ear microphone based on the BOYA BY-M1 Lavalier microphone is recommended.

This project is a part of my graduation work, Cyrillic symbols are one of requirements for the system.