DeepFry: Identifying Vocal Fry Using Deep Neural Networks

TL;DR

This paper introduces a deep learning model that accurately detects vocal fry in speech, addressing challenges posed by irregular glottal vibrations and improving recognition systems for languages with prevalent creaky voice.

Contribution

It presents a novel encoder-classifier deep neural network that learns from raw waveforms and refines creak detection using auxiliary voice features, outperforming previous methods.

Findings

Improved recall and F1 scores on unseen data

Effective use of raw waveform and auxiliary features

Enhanced detection of creaky voice in American English

Abstract

Vocal fry or creaky voice refers to a voice quality characterized by irregular glottal opening and low pitch. It occurs in diverse languages and is prevalent in American English, where it is used not only to mark phrase finality, but also sociolinguistic factors and affect. Due to its irregular periodicity, creaky voice challenges automatic speech processing and recognition systems, particularly for languages where creak is frequently used. This paper proposes a deep learning model to detect creaky voice in fluent speech. The model is composed of an encoder and a classifier trained together. The encoder takes the raw waveform and learns a representation using a convolutional neural network. The classifier is implemented as a multi-headed fully-connected network trained to detect creaky voice, voicing, and pitch, where the last two are used to refine creak prediction. The model is trained and tested on speech of American English speakers, annotated for creak by trained phoneticians. We evaluated the performance of our system using two encoders: one is tailored for the task, and the other is based on a state-of-the-art unsupervised representation. Results suggest our best-performing system has improved recall and F1 scores compared to previous methods on unseen data.