Crate axonml_audio 

Axonml Audio - Audio Processing Utilities

This crate provides audio processing functionality for the Axonml ML framework:

• Transforms: Resampling, spectrograms, MFCC, pitch shifting, noise addition
• Datasets: Audio classification, command recognition, music genre, speaker ID

Example

use axonml_audio::prelude::*;

// Create a synthetic command dataset
let dataset = SyntheticCommandDataset::small();

// Apply mel spectrogram transform
let mel = MelSpectrogram::new(16000);
let (waveform, label) = dataset.get(0).unwrap();
let spectrogram = mel.apply(&waveform);

@version 0.1.0 @author AutomataNexus Development Team

Re-exports

pub use transforms::AddNoise;

pub use transforms::MelSpectrogram;

pub use transforms::NormalizeAudio;

pub use transforms::PitchShift;

pub use transforms::Resample;

pub use transforms::TimeStretch;

pub use transforms::TrimSilence;

pub use transforms::MFCC;

pub use datasets::AudioClassificationDataset;

pub use datasets::AudioSeq2SeqDataset;

pub use datasets::SyntheticCommandDataset;

pub use datasets::SyntheticMusicDataset;

pub use datasets::SyntheticSpeakerDataset;

Modules

datasets

Audio Datasets - Dataset implementations for audio processing tasks

prelude

Common imports for audio processing.

transforms

Audio Transforms - Signal Processing and Augmentation