Improving Unsupervised Subword Modeling via Disentangled Speech Representation Learning and Transformation

2019-06-17 · Siyuan Feng, Tan Lee

This study tackles unsupervised subword modeling in the zero-resource scenario, learning frame-level speech representation that is phonetically discriminative and speaker-invariant, using only untranscribed speech for target languages. Frame label acquisition is an essential step in solving this problem. High quality frame labels should be in good consistency with golden transcriptions and robust to speaker variation. We propose to improve frame label acquisition in our previously adopted deep neural network-bottleneck feature (DNN-BNF) architecture by applying the factorized hierarchical variational autoencoder (FHVAE). FHVAEs learn to disentangle linguistic content and speaker identity information encoded in speech. By discarding or unifying speaker information, speaker-invariant features are learned and fed as inputs to DPGMM frame clustering and DNN-BNF training. Experiments conducted on ZeroSpeech 2017 show that our proposed approaches achieve $2.4\%$ and $0.6\%$ absolute ABX error rate reductions in across- and within-speaker conditions, comparing to the baseline DNN-BNF system without applying FHVAEs. Our proposed approaches significantly outperform vocal tract length normalization in improving frame labeling and subword modeling.

📄 PDF Abstract BibTeX arXiv:1906.07245

Code (0)

등록된 구현이 없습니다.

Tasks

ClusteringRepresentation LearningSpeech Representation Learning

Methods 이 논문이 사용한 방법론

Solana Customer Service Number +1-833-534-1729 설명 없음

Combining Adversarial Training and Disentangled Speech Representation for Robust Zero-Resource Subword Modeling

2019-06-17 · Siyuan Feng, Tan Lee, Zhiyuan Peng

This study addresses the problem of unsupervised subword unit discovery from untranscribed speech. It forms the basis of the ultimate goal of ZeroSpeech 2019, building text-to-speech systems without text labels. In this …

Representation Learning Speech Representation Learning text-to-speech Text to Speech

Exploiting Cross-Lingual Knowledge in Unsupervised Acoustic Modeling for Low-Resource Languages

2020-07-29 · Siyuan Feng

(Short version of Abstract) This thesis describes an investigation on unsupervised acoustic modeling (UAM) for automatic speech recognition (ASR) in the zero-resource scenario, where only untranscribed speech data is ass…

Automatic Speech Recognition Automatic Speech Recognition (ASR)Language Acquisition speech-recognition+1

Exploiting Cross-Lingual Speaker and Phonetic Diversity for Unsupervised Subword Modeling

2019-08-09 · Siyuan Feng, Tan Lee

This research addresses the problem of acoustic modeling of low-resource languages for which transcribed training data is absent. The goal is to learn robust frame-level feature representations that can be used to identi…

Automatic Speech Recognition Automatic Speech Recognition (ASR)Clustering Diversity+3

Confusion2vec 2.0: Enriching Ambiguous Spoken Language Representations with Subwords

2021-02-03 · Prashanth Gurunath Shivakumar, Panayiotis Georgiou, Shrikanth Narayanan

Word vector representations enable machines to encode human language for spoken language understanding and processing. Confusion2vec, motivated from human speech production and perception, is a word vector representation…

Automatic Speech Recognition Automatic Speech Recognition (ASR)Intent Detection Natural Language Understanding+4

Unsupervised Acoustic Unit Discovery by Leveraging a Language-Independent Subword Discriminative Feature Representation

2021-04-02 · Siyuan Feng, Piotr Żelasko, Laureano Moro-Velázquez, Odette Scharenborg

This paper tackles automatically discovering phone-like acoustic units (AUD) from unlabeled speech data. Past studies usually proposed single-step approaches. We propose a two-stage approach: the first stage learns a sub…

Acoustic Unit Discovery Clustering