Publications associées à On dynamic stream weighting for Audio-Visual Speech Recognition

Post-correction of Historical Text Transcripts with Large Language Models: An Exploratory Study

Frédéric Kaplan, Maud Ehrmann, Matteo Romanello, Sven-Nicolas Yoann Najem, Emanuela Boros

The quality of automatic transcription of heritage documents, whether from printed, manuscripts or audio sources, has a decisive impact on the ability to search and process historical texts. Although significant progress has been made in text recognition ( ...

Association for Computational Linguistics2024

Sparse Autoencoders for Speech Modeling and Recognition

Selen Hande Kabil

Speech recognition-based applications upon the advancements in artificial intelligence play an essential role to transform most aspects of modern life. However, speech recognition in real-life conditions (e.g., in the presence of overlapping speech, varyin ...

EPFL2023

Novel Methods For Detection And Analysis Of Atypical Aspects In Speech

Julian David Fritsch

Atypical aspects in speech concern speech that deviates from what is commonly considered normal or healthy. In this thesis, we propose novel methods for detection and analysis of these aspects, e.g. to monitor the temporary state of a speaker, diseases tha ...

EPFL2023

Explanation of Face Recognition via Saliency Maps

Touradj Ebrahimi, Yuhang Lu

Despite the significant progress in recent years, deep face recognition is often treated as a "black box" and has been criticized for lacking explainability. It becomes increasingly important to understand the characteristics and decisions of deep face rec ...

2023

On matching data and model in LF-MMI-based dysarthric speech recognition

Enno Hermann

In light of steady progress in machine learning, automatic speech recognition (ASR) is entering more and more areas of our daily life, but people with dysarthria and other speech pathologies are left behind. Their voices are underrepresented in the trainin ...

EPFL2023

A Virtual Simulation-Pilot Agent for Training of Air Traffic Controllers

Petr Motlicek, Amrutha Prasad, Juan Pablo Zuluaga Gomez

In this paper we propose a novel virtual simulation-pilot engine for speeding up air traffic controller (ATCo) training by integrating different state-of-the-art artificial intelligence (AI)-based tools. The virtual simulation-pilot engine receives spoken ...

MDPI2023

Observation of non-reciprocal harmonic conversion in real sounds

Hervé Lissek, Romain Christophe Rémy Fleury, Xinxin Guo

Reciprocity guarantees that in most media, sound transmission is symmetric between two points of space when the location of the source and receiver are interchanged. This fundamental law can be broken in non-linear media, often at the cost of detrimental i ...

2023

The Impact of Data Persistence Bias on Social Media Studies

Tugrulcan Elmas

Social media studies often collect data retrospectively to analyze public opinion. Social media data may decay over time and such decay may prevent the collection of the complete dataset. As a result, the collected dataset may differ from the complete data ...

Assoc Computing Machinery2023

Can Self-Supervised Neural Representations Pre-Trained on Human Speech distinguish Animal Callers?

Mathew Magimai Doss, Eklavya Sarkar

Self-supervised learning (SSL) models use only the intrinsic structure of a given signal, independent of its acoustic domain, to extract essential information from the input to an embedding space. This implies that the utility of such representations is no ...

ISCA2023

Lausanne Historical Censuses Dataset HTR 35k

Marion Kramer, Rémi Guillaume Petitpierre, Lucas Arnaud André Rappo

This training dataset includes a total of 34,913 manually transcribed text segments. It is dedicated to the handwritten text recognition (HTR) of historical sources, typically tabular records, such as censuses. This dataset is based on a sample of 83 pages ...

Zenodo2023

How to Boost Face Recognition with StyleGAN?

Nikita Durasov

State-of-the-art face recognition systems require vast amounts of labeled training data. Given the priority of privacy in face recognition applications, the data is limited to celebrity web crawls, which have issues such as limited numbers of identities. O ...

Ieee Computer Soc2023

Johann Sebastian Bach – The Chorales

Martin Alois Rohrmeier, Johannes Hentschel

This is a README file for a data repository originating from the DCML corpus initiative and serves as welcome page for both the GitHub repo

Zenodo2023

Recognizing distant faces

Marin Vogelsang, Lukas Vogelsang

As an 'early alerting' sense, one of the primary tasks for the human visual system is to recognize distant objects. In the specific context of facial identification, this ecologically important task has received surprisingly little attention. Most studies ...

PERGAMON-ELSEVIER SCIENCE LTD2023

Unsupervised Voice Activity Detection by Modeling Source and System Information using Zero Frequency Filtering

Mathew Magimai Doss, Eklavya Sarkar

Voice activity detection (VAD) is an important pre-processing step for speech technology applications. The task consists of deriving segment boundaries of audio signals which contain voicing information. In recent years, it has been shown that voice source ...

ISCA2022

On Breathing Pattern Information in Synthetic Speech

Mathew Magimai Doss, Zohreh Mostaani

The respiratory system is an integral part of human speech production. As a consequence, there is a close relation between respiration and speech signal, and the produced speech signal carries breathing pattern related information. Speech can also be gener ...

ISCA-INT SPEECH COMMUNICATION ASSOC2022

Automatic pathological speech assessment

Parvaneh Janbakhshi

Many pathologies cause impairments in the speech production mechanism resulting in reduced speech intelligibility and communicative ability. To assist the clinical diagnosis, treatment and management of speech disorders, automatic pathological speech asses ...

EPFL2022

A Functional Perspective on Information Measures

Amedeo Roberto Esposito

Since the birth of Information Theory, researchers have defined and exploited various information measures, as well as endowed them with operational meanings. Some were born as a "solution to a problem", like Shannon's Entropy and Mutual Information. Other ...

EPFL2022

A Two-Step Approach To Leverage Contextual Data: Speech Recognition In Air-Traffic Communications

Petr Motlicek, Amrutha Prasad, Juan Pablo Zuluaga Gomez

Automatic Speech Recognition (ASR), as the assistance of speech communication between pilots and air-traffic controllers, can significantly reduce the complexity of the task and increase the reliability of transmitted information. ASR application can lead ...

IEEE2022

Nothing New Under the Sun: Novelty Constructs and Measures in Social Studies

Davide Bavato

The concept of novelty is central to questions of creativity, innovation, and discovery. Despite the prominence in scientific inquiry and everyday discourse, there is a chronic ambiguity over its meaning and a surprising variety of empirical measures, whic ...

Emerald Publishing Limited2022

Bertraffic: Bert-Based Joint Speaker Role And Speaker Change Detection For Air Traffic Control Communications

Petr Motlicek, Amrutha Prasad, Juan Pablo Zuluaga Gomez

Automatic speech recognition (ASR) allows transcribing the communications between air traffic controllers (ATCOs) and aircraft pilots. The transcriptions are used later to extract ATC named entities, e.g., aircraft callsigns. One common challenge is speech ...

IEEE2022