English
Related papers

Related papers: Opening the Black Box of wav2vec Feature Encoder

200 papers

Learning good representations is of crucial importance in deep learning. Mutual Information (MI) or similar measures of statistical dependence are promising tools for learning these representations in an unsupervised way. Even though the…

Audio and Speech Processing · Electrical Eng. & Systems 2019-04-09 Mirco Ravanelli , Yoshua Bengio

Embedding acoustic information into fixed length representations is of interest for a whole range of applications in speech and audio technology. Two novel unsupervised approaches to generate acoustic embeddings by modelling of acoustic…

Computation and Language · Computer Science 2021-02-08 Yanpei Shi , Thomas Hain

The aim of latent variable disentanglement is to infer the multiple informative latent representations that lie behind a data generation process and is a key factor in controllable data generation. In this paper, we propose a deep neural…

Sound · Computer Science 2023-09-07 Yiming Wu

Autoencoders are a widespread tool in machine learning to transform high-dimensional data into a lowerdimensional representation which still exhibits the essential characteristics of the input. The encoder provides an embedding from the…

Machine Learning · Computer Science 2021-04-28 Juliane Braunsmann , Marko Rajković , Martin Rumpf , Benedikt Wirth

Neural latent variable models enable the discovery of interesting structure in speech audio data. This paper presents a comparison of two different approaches which are broadly based on predicting future time-steps or auto-encoding the…

Audio and Speech Processing · Electrical Eng. & Systems 2020-10-28 Henry Zhou , Alexei Baevski , Michael Auli

Automatic Speech Recognition has advanced with self-supervised learning, enabling feature extraction directly from raw audio. In Wav2Vec, a CNN first transforms audio into feature vectors before the transformer processes them. This study…

Computation and Language · Computer Science 2025-08-26 Domenico De Cristofaro , Vincenzo Norman Vitale , Alessandro Vietti

Given the speech generation framework that represents the speaker attribute with an embedding vector, asynchronous voice anonymization can be achieved by modifying the speaker embedding derived from the original speech. However, the…

Audio and Speech Processing · Electrical Eng. & Systems 2025-10-08 Rui Wang , Liping Chen , Kong Aik Lee , Zhengpeng Zha , Zhenhua Ling

Speech representation and modelling in high-dimensional spaces of acoustic waveforms, or a linear transformation thereof, is investigated with the aim of improving the robustness of automatic speech recognition to additive noise. The…

Computation and Language · Computer Science 2015-03-31 Matthew Ager , Zoran Cvetkovic , Peter Sollich

Recent studies have shown that sequence-to-sequence (seq2seq) models struggle with compositional generalization (CG), i.e., the ability to systematically generalize to unseen compositions of seen components. There is mounting evidence that…

Computation and Language · Computer Science 2023-10-19 Lei Lin , Shuangtao Li , Yafang Zheng , Biao Fu , Shan Liu , Yidong Chen , Xiaodong Shi

Variational encoder-decoders (VEDs) have shown promising results in dialogue generation. However, the latent variable distributions are usually approximated by a much simpler model than the powerful RNN structure used for encoding and…

Computation and Language · Computer Science 2018-02-07 Xiaoyu Shen , Hui Su , Shuzi Niu , Vera Demberg

The expressive nature of the voice provides a powerful medium for communicating sonic ideas, motivating recent research on methods for query by vocalisation. Meanwhile, deep learning methods have demonstrated state-of-the-art results for…

Multimedia · Computer Science 2018-02-15 Adib Mehrabi , Keunwoo Choi , Simon Dixon , Mark Sandler

The articulatory geometric configurations of the vocal tract and the acoustic properties of the resultant speech sound are considered to have a strong causal relationship. This paper aims at finding a joint latent representation between the…

Audio and Speech Processing · Electrical Eng. & Systems 2020-10-02 Pramit Saha , Sidney Fels

Speech 'in-the-wild' is a handicap for speaker recognition systems due to the variability induced by real-life conditions, such as environmental noise and the emotional state of the speaker. Taking advantage of the principles of…

Audio and Speech Processing · Electrical Eng. & Systems 2022-05-17 Esther Rituerto-González , Carmen Peláez-Moreno

How do speech models trained through self-supervised learning structure their representations? Previous studies have looked at how information is encoded in feature vectors across different layers. But few studies have considered whether…

Audio and Speech Processing · Electrical Eng. & Systems 2026-05-11 Kyle Janse van Rensburg , Benjamin van Niekerk , Herman Kamper

It is increasingly considered that human speech perception and production both rely on articulatory representations. In this paper, we investigate whether this type of representation could improve the performances of a deep generative model…

Sound · Computer Science 2021-04-08 Marc-Antoine Georges , Laurent Girin , Jean-Luc Schwartz , Thomas Hueber

Pre-trained deep learning embeddings have consistently shown superior performance over handcrafted acoustic features in speech emotion recognition (SER). However, unlike acoustic features with clear physical meaning, these embeddings lack…

Sound · Computer Science 2024-09-17 Satvik Dixit , Daniel M. Low , Gasser Elbanna , Fabio Catania , Satrajit S. Ghosh

Timbre is a set of perceptual attributes that identifies different types of sound sources. Although its definition is usually elusive, it can be seen from a signal processing viewpoint as all the spectral features that are perceived…

Audio and Speech Processing · Electrical Eng. & Systems 2020-07-14 Adrien Bitton , Philippe Esling , Tatsuya Harada

Phonetic normalization plays a crucial role in speech recognition and analysis, ensuring the comparability of features derived from raw audio data. However, in the current paradigm of fine-tuning pre-trained large transformer models,…

Computation and Language · Computer Science 2025-03-10 Yiming Wang , Yi Yang , Jiahong Yuan

Unsupervised representation learning of speech has been of keen interest in recent years, which is for example evident in the wide interest of the ZeroSpeech challenges. This work presents a new method for learning frame level…

Audio and Speech Processing · Electrical Eng. & Systems 2020-08-18 Mingjie Chen , Thomas Hain

Recent advancements in neural audio codecs have not only enabled superior audio compression but also enhanced speech synthesis techniques. Researchers are now exploring their potential as universal acoustic feature extractors for a broader…

Audio and Speech Processing · Electrical Eng. & Systems 2025-11-21 Wei-Cheng Tseng , David Harwath