Unsupervised Representation Learning Of Speech For Dialect Identification

Abstract

In this paper, we explore the use of a factorized hierarchical variational autoencoder (FHVAE) model to learn an unsupervised latent representation for dialect identification (DID). An FHVAE can learn a latent space that separates the more static attributes within an utterance from the more dynamic attributes by encoding them into two different sets of latent variables. Useful factors for dialect identification, such as phonetic or linguistic content, are encoded by a segmental latent variable, while irrelevant factors that are relatively constant within a sequence, such as a channel or a speaker information, are encoded by a sequential latent variable. The disentanglement property makes the segmental latent variable less susceptible to channel and speaker variation, and thus reduces degradation from channel domain mismatch. We demonstrate that on fully-supervised DID tasks, an end-to-end model trained on the features extracted from the FHVAE model achieves the best performance, compar

Stats

• citations8
• S2 citations—
• github stars0
• HF likes0
• heat score7.16
• arxiv keyshon2018unsupervised

Related papers

• Improved Disentangled Speech Representations Using Contrastive Learning In Factorized Hierarchical Variational Autoencoder (2022)2.26
• Disentangled Speech Representation Learning Based On Factorized Hierarchical Variational Autoencoder With Self-supervised Objective (2022)7.81
• Weak-supervised Dysarthria-invariant Features For Spoken Language Understanding Using An FHVAE And Adversarial Training (2022)2.26
• Improving Unsupervised Subword Modeling Via Disentangled Speech Representation Learning And Transformation (2019)5.24
• Extracting Domain Invariant Features By Unsupervised Learning For Robust Automatic Speech Recognition (2018)9.03
• Mixture Factorized Auto-encoder For Unsupervised Hierarchical Deep Factorization Of Speech Signal (2019)0.00
• Variational Auto-encoder Based Variability Encoding For Dysarthric Speech Recognition (2022)7.16
• Multi-view Dimensionality Reduction For Dialect Identification Of Arabic Broadcast Speech (2016)0.00