TranscribeASR

TranscribeASR (TASR, pronounced "Tay-ser") is a simple CLI based on ASR technology, focused on quickly and accurately converting audio files into text, helping users improve their work efficiency.

Installation

Install via pip:

pip install tasr

Dependencies

Install dependencies:

pip install -r requirements.txt

Install Demucs (Optional, only required for noise reduction):

pip install demucs

Usage

Basic Usage

tasr --input <audio_file_path> [--output <output_text_file_path>] [--language <language>] [--denoise]

Examples

Basic Speech Recognition

Perform speech recognition on the audio file example.wav and save the result as example.txt:

tasr --input example.wav

Specify Output File

Perform speech recognition on the audio file example.wav and save the result as output.txt:

python cli.py --input example.wav --output output.txt

Notes

Supported audio formats depend on the capabilities of soundfile and demucs, typically including .wav, .flac, .mp3, etc. For best results with noise reduction, it is recommended to use .wav format.