Convolutional Variational Autoencoders for Spectrogram Compression in Automatic Speech Recognition

TL;DR

This paper introduces a convolutional variational autoencoder approach to compress spectrograms for speech recognition, enabling efficient feature extraction that improves ASR performance over traditional MFCC features.

Contribution

It presents a novel convolutional VAE method for spectrogram compression and demonstrates its effectiveness in speech recognition tasks.

Findings

VAE-based features outperform MFCC in ASR accuracy

Compressed spectrogram representations reduce feature dimensionality

Model successfully generalizes to spoken command datasets

Abstract

For many Automatic Speech Recognition (ASR) tasks audio features as spectrograms show better results than Mel-frequency Cepstral Coefficients (MFCC), but in practice they are hard to use due to a complex dimensionality of a feature space. The following paper presents an alternative approach towards generating compressed spectrogram representation, based on Convolutional Variational Autoencoders (VAE). A Convolutional VAE model was trained on a subsample of the LibriSpeech dataset to reconstruct short fragments of audio spectrograms (25 ms) from a 13-dimensional embedding. The trained model for a 40-dimensional (300 ms) embedding was used to generate features for corpus of spoken commands on the GoogleSpeechCommands dataset. Using the generated features an ASR system was built and compared to the model with MFCC features.