#speech-recognition #offline #next-generation #open #source #wav #asr

april_asr

Offline open source speech recognition API based on next-generation Kaldi

4 releases

0.1.31 Jun 5, 2023
0.1.21-dev2 Jun 5, 2023

#425 in Audio

GPL-3.0 license

37KB
773 lines

april_asr

This crate is a (safe) wrapper around April ASR through its C API. From its GitHub page:

  • April-ASR is a minimal library that provides an API for offline streaming speech-to-text applications

Usage

Reading a WAV file and feeding it to the APRIL-ASR model:

use april_asr::{Model, Session, Result, AprilResult};
use std::io::Read;

april_asr::april_asr_init();

let model = Model::new("/path/to/model.april").unwrap();

let session = Session::new(model, april_asr::SessionFlags::SYNC, |result: AprilResult| {
    println!("Result: {:?}", result);
});
let mut buffer = Vec::new();
let mut file = std::fs::File::open("/path/to/file.wav").unwrap();
let _ = file.read(&mut buffer);
session.feed_pcm16(buffer);

session.flush();

Dependencies

~0–3MB
~54K SLoC