40 citations · 85 across the 23 of their papers we have counts for
34 papers
Regularizing Contrastive Predictive Coding for Speech Applications
Saurabhchand Bhati, Jesús Villalba, Piotr Żelasko +2
Self-supervised methods such as Contrastive predictive Coding (CPC) have greatly improved the quality of the unsupervised representations. These representations significantly reduc…
Predicting Multi-Codebook Vector Quantization Indexes for Knowledge Distillation
Liyong Guo, Xiaoyu Yang, Quandong Wang +9
Knowledge distillation(KD) is a common approach to improve model performance in automatic speech recognition (ASR), where a student model is trained to imitate the output behaviour…
Delay-penalized transducer for low-latency streaming ASR
Wei Kang, Zengwei Yao, Fangjun Kuang +5
In streaming automatic speech recognition (ASR), it is desirable to reduce latency as much as possible while having minimum impact on recognition accuracy. Although a few existing…
Fast and parallel decoding for transducer
Wei Kang, Liyong Guo, Fangjun Kuang +6
The transducer architecture is becoming increasingly popular in the field of speech recognition, because it is naturally streaming as well as high in accuracy. One of the drawbacks…
Time-domain speech super-resolution with GAN based modeling for telephony speaker verification
Saurabh Kataria, Jesús Villalba, Laureano Moro-Velázquez +2
Automatic Speaker Verification (ASV) technology has become commonplace in virtual assistants. However, its performance suffers when there is a mismatch between the train and test d…
Non-Contrastive Self-Supervised Learning of Utterance-Level Speech Representations
Jaejin Cho, Raghavendra Pappagari, Piotr Żelasko +3
Considering the abundance of unlabeled speech data and the high labeling costs, unsupervised learning methods can be essential for better system development. One of the most succes…
Defense against Adversarial Attacks on Hybrid Speech Recognition using Joint Adversarial Fine-tuning with Denoiser
Sonal Joshi, Saurabh Kataria, Yiwen Shao +4
Adversarial attacks are a threat to automatic speech recognition (ASR) systems, and it becomes imperative to propose defenses to protect them. In this paper, we perform experiments…
Discovering Phonetic Inventories with Crosslingual Automatic Speech Recognition
Piotr Żelasko, Siyuan Feng, Laureano Moro Velazquez +5
The high cost of data acquisition makes Automatic Speech Recognition (ASR) model training problematic for most existing languages, including languages that do not even have a writt…
Lhotse: a speech data representation library for the modern deep learning ecosystem
Piotr Żelasko, Daniel Povey, Jan "Yenda" Trmal +1
Speech data is notoriously difficult to work with due to a variety of codecs, lengths of recordings, and meta-data formats. We present Lhotse, a speech data representation library…
Unsupervised Speech Segmentation and Variable Rate Representation Learning using Segmental Contrastive Predictive Coding
Saurabhchand Bhati, Jesús Villalba, Piotr Żelasko +2
Typically, unsupervised segmentation of speech into the phone and word-like units are treated as separate tasks and are often done via different methods which do not fully leverage…
Beyond Isolated Utterances: Conversational Emotion Recognition
Raghavendra Pappagari, Piotr Żelasko, Jesús Villalba +2
Speech emotion recognition is the task of recognizing the speaker's emotional state given a recording of their utterance. While most of the current approaches focus on inferring em…
Joint prediction of truecasing and punctuation for conversational speech in low-resource scenarios
Raghavendra Pappagari, Piotr Żelasko, Agnieszka Mikołajczyk +2
Capitalization and punctuation are important cues for comprehending written texts and conversational transcripts. Yet, many ASR systems do not produce punctuated and case-formatted…
Representation Learning to Classify and Detect Adversarial Attacks against Speaker and Speech Recognition Systems
Jesús Villalba, Sonal Joshi, Piotr Żelasko +1
Adversarial attacks have become a major threat for machine learning applications. There is a growing interest in studying these attacks in the audio domain, e.g, speech and speaker…
What Helps Transformers Recognize Conversational Structure? Importance of Context, Punctuation, and Labels in Dialog Act Recognition
Piotr Żelasko, Raghavendra Pappagari, Najim Dehak
Dialog acts can be interpreted as the atomic units of a conversation, more fine-grained than utterances, characterized by a specific communicative function. The ability to structur…
Segmental Contrastive Predictive Coding for Unsupervised Word Segmentation
Saurabhchand Bhati, Jesús Villalba, Piotr Żelasko +2
Automatic detection of phoneme or word-like units is one of the core objectives in zero-resource speech processing. Recent attempts employ self-supervised training methods, such as…
Deep Feature CycleGANs: Speaker Identity Preserving Non-parallel Microphone-Telephone Domain Adaptation for Speaker Verification
Saurabh Kataria, Jesús Villalba, Piotr Żelasko +2
With the increase in the availability of speech from varied domains, it is imperative to use such out-of-domain data to improve existing speech systems. Domain adaptation is a prom…
Earnings-21: A Practical Benchmark for ASR in the Wild
Miguel Del Rio, Natalie Delworth, Ryan Westerman +7
Commonly used speech corpora inadequately challenge academic and commercial ASR systems. In particular, speech corpora lack metadata needed for detailed analysis and WER measuremen…
Unsupervised Acoustic Unit Discovery by Leveraging a Language-Independent Subword Discriminative Feature Representation
Siyuan Feng, Piotr Żelasko, Laureano Moro-Velázquez +1
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-stag…
Adversarial Attacks and Defenses for Speech Recognition Systems
Piotr Żelasko, Sonal Joshi, Yiwen Shao +4
The ubiquitous presence of machine learning systems in our lives necessitates research into their vulnerabilities and appropriate countermeasures. In particular, we investigate the…
Study of Pre-processing Defenses against Adversarial Attacks on State-of-the-art Speaker Recognition Systems
Sonal Joshi, Jesús Villalba, Piotr Żelasko +2
Adversarial examples to speaker recognition (SR) systems are generated by adding a carefully crafted noise to the speech signal to make the system fail while being imperceptible to…
Focus on the present: a regularization method for the ASR source-target attention layer
Nanxin Chen, Piotr Żelasko, Jesús Villalba +1
This paper introduces a novel method to diagnose the source-target attention in state-of-the-art end-to-end speech recognition models with joint connectionist temporal classificati…
Learning Speaker Embedding from Text-to-Speech
Jaejin Cho, Piotr Zelasko, Jesus Villalba +2
Zero-shot multi-speaker Text-to-Speech (TTS) generates target speaker voices given an input text and the corresponding speaker embedding. In this work, we investigate the effective…
WER we are and WER we think we are
Piotr Szymański, Piotr Żelasko, Mikolaj Morzy +6
Natural language processing of conversational speech requires the availability of high-quality transcripts. In this paper, we express our skepticism towards the recent reports of v…
CopyPaste: An Augmentation Method for Speech Emotion Recognition
Raghavendra Pappagari, Jesús Villalba, Piotr Żelasko +2
Data augmentation is a widely used strategy for training robust machine learning models. It partially alleviates the problem of limited data for tasks like speech emotion recogniti…