Cost-effective, Multilingual, Privacy-driven voice-enabled Services

Livrables

First prototype integrating the research results of WP2 WP3 and T41

Final platform demonstrator and updated data protection and GDPR requirements

Final platform demonstrator populated with additional data and trained models Updated version of D51 taking the latest research and legal advances into account

Initial platform demonstrator

Fully functional platform demonstrator implemented populated with a few initial data and trained models deployed in working environment ready for use in other WPs

Final COMPRISE SDK prototype and documentation

Final prototype integrating the research results of WP2 WP3 and T41 and Swagger online documentation

Second dissemination and communication report

Update on the actions conducted lessons learned on the innovation ecosystem proposed in COMPRISE and planned dissemination and communication after the end of the project

Dissemination and communication action plan

Summarises all planned dissemination actions and provides communication material to the partners (logo, graphical chart, public website, short presentation of the project, templates, poster, leaflet)

First dissemination and communication report

Update on the actions conducted and web document defining the COMPRISE knowledge repository on multilingual voice-enabled applications.

Final weakly supervised learning library

Final design implementation and evaluation of weakly supervised learning for all considered tasks

Baseline speech and text transformation and model learning library

Design, implementation, and evaluation of baseline transformations focusing on deleting the user’s identity and words carrying critical information, and model learning.

Improved transformation library and initial privacy guarantees

Design, implementation, and evaluation of speech and text transformations addressing more types of private information and initial statistical utility/privacy bounds.

Final personalised learning library

Final design implementation and evaluation of model personalisation strategies for speechtotext spoken language understanding and dialog management

Initial multilingual interaction library

Software components and documentation for speech-to-speech translation and integration of dialog systems in the operating branch.

Final transformation library and privacy guarantees

Final design implementation and evaluation of speech and text transformations and final statistical utilityprivacy bounds

Data collection and curation features of the platform

Platform with data collection and curation features implemented and deployed in working environment

Final multilingual interaction library

Software components and documentation for speechtospeech translation and integration of dialog systems in both the operating and the training branch

Initial weakly supervised learning library

Design, implementation, and evaluation of weakly supervised learning for spoken language understanding.

Initial personalised learning library for speech-to-text

Design, implementation, and evaluation of initial model personalisation strategies for speech-to-text.

Initial data management plan
Final data management plan

Initial scientific evaluation

First combined evaluation of baseline speech, dialog, and translation tools.

Platform hardware and software architecture

Platform specification including main requirements, hardware and software architecture

SDK software architecture

SDK specification including main requirements and software architecture

Data protection and GDPR requirements

Guidelines, procedures and recommendations how to implement personal data protection in the platform.

Publications

Transfer Learning and Distant Supervision for Multilingual Transformer Models: A Study on African Languages

Auteurs: Michael A. Hedderich; David Ifeoluwa Adelani; Dawei Zhu; Jesujoba O. Alabi; Udia Markus; Dietrich Klakow
Publié dans: 2020 Conference on Empirical Methods in Natural Language Processing, Numéro 16/11/2020, 2020
Éditeur: ACL
DOI: 10.18653/v1/2020.emnlp-main.204

Distant supervision and noisy label learning for low resource named entity recognition: A study on Hausa and Yorùbá

Auteurs: Adelani, David Ifeoluwa; Hedderich, Michael,; Zhu, Dawei; Van Den Berg, Esther; Klakow, Dietrich
Publié dans: AfricaNLP / PML4DC Workshop 2020 @ICLR 2020, Numéro 26/04/2020, 2020
Éditeur: OpenReview.net

Preventing author profiling through zero-shot multilingual back-translation

Auteurs: Adelani, David,; Zhang, Miaoran; Shen, Xiaoyu; Davody, Ali; Kleinbauer, Thomas; Klakow, Dietrich
Publié dans: 2021 Conference on Empirical Methods in Natural Language Processing, Numéro 07/11/2021, 2021
Éditeur: ACL

The effect of domain and diacritics in Yorùbá-English neural machine translation

Auteurs: Adelani, David,; Ruiter, Dana; Alabi, Jesujoba,; Adebonojo, Damilola; Ayeni, Adesina; Adeyemi, Mofetoluwa; Awokoya, Ayodele; Espana-Bonet, Cristina
Publié dans: 18th Biennial Machine Translation Summit, Numéro 16/08/2021, 2021
Éditeur: AMTA

Benchmarking and challenges in security and privacy for voice biometrics

Auteurs: Bonastre, Jean-Francois; Delgado, Hector; Evans, Nicholas; Kinnunen, Tomi; Lee, Kong Aik; Liu, Xuechen; Nautsch, Andreas; Noe, Paul-Gauthier; Patino, Jose; Sahidullah, Md; Srivastava, Brij Mohan Lal; Todisco, Massimiliano; Tomashenko, Natalia; Vincent, Emmanuel; Wang, Xin; Yamagishi, Junichi
Publié dans: 1st ISCA Symposium on Security and Privacy in Speech Communication, Numéro 10/11/2021, 2021
Éditeur: ISCA

Privacy-Preserving Adversarial Representation Learning in ASR: Reality or Illusion?

Auteurs: Srivastava, Brij Mohan Lal; Bellet, Aurélien; Tommasi, Marc; Vincent, Emmanuel
Publié dans: INTERSPEECH 2019, Numéro 15/09/2019, 2019, Page(s) 3700-3704
Éditeur: ISCA

Introducing the VoicePrivacy initiative

Auteurs: Tomashenko, Natalia; Srivastava, Brij Mohan Lal,; Wang, Xin; Vincent, Emmanuel; Nautsch, Andreas; Yamagishi, Junichi; Evans, Nicholas; Patino, Jose; Bonastre, J.-F; Noé, Paul-Gauthier; Todisco, Massimiliano
Publié dans: INTERSPEECH 2020, Numéro 25/10/2020, 2020
Éditeur: ISCA

Evaluating Voice Conversion-based Privacy Protection against Informed Attackers

Auteurs: Srivastava, Brij Mohan Lal; Vauquier, Nathalie; Sahidullah, Md; Bellet, Aurélien; Tommasi, Marc; Vincent, Emmanuel
Publié dans: 2020 IEEE International Conference on Acoustics, Speech, and Signal Processing, Numéro 04/05/2020, 2020, Page(s) 2802-2806
Éditeur: IEEE

Investigating the Impact of Pre-trained Word Embeddings on Memorization in Neural Networks

Auteurs: Thomas, Aleena; Adelani, David; Davody, Ali; Mogadala, Aditya; Klakow, Dietrich
Publié dans: 23rd International Conference on Text, Speech and Dialogue, Numéro 08/09/2020, 2020
Éditeur: Springer

Design Choices for X-vector Based Speaker Anonymization

Auteurs: Srivastava, Brij Mohan Lal; Tomashenko, Natalia; Wang, Xin; Vincent, Emmanuel; Yamagishi, Junichi; Maouche, Mohamed; Bellet, Aurélien; Tommasi, Marc
Publié dans: INTERSPEECH 2020, Numéro 25/10/2020, 2020
Éditeur: ISCA

The COMPRISE Cloud Platform

Auteurs: Skadiņš, Raivis; Salimbajevs, Askars
Publié dans: 1st International Workshop on Language Technology Platforms, Numéro 16/05/2020, 2020, Page(s) 108–111
Éditeur: European Language Resources Association

Using Privacy-Transformed Speech in the Automatic Speech Recognition Acoustic Model Training

Auteurs: Salimbajevs, Askars
Publié dans: 9th International Conference on Human Language Technologies – the Baltic Perspective, Numéro 22/09/2020, 2020
Éditeur: IOS Press

Assessing Unintended Memorization in Neural Discriminative Sequence Models

Auteurs: Helali, Mossad; Kleinbauer, Thomas; Klakow, Dietrich
Publié dans: 23rd International Conference on Text, Speech and Dialogue, Numéro 08/09/2020, 2020
Éditeur: Springer

Private Protocols for U-Statistics in the Local Model and Beyond

Auteurs: Bell, James; Bellet, Aurélien; Gascón, Adrià; Kulkarni, Tejas
Publié dans: International Conference on Artificial Intelligence and Statistics, Numéro 27/08/2020, 2020, Page(s) 1573-1583
Éditeur: Proceedings of Machine Learning Research

Data Augmentation for Pipeline-Based Speech Translation

Auteurs: Alves, Diego; Salimbajevs, Askars; Pinnis, Mārcis
Publié dans: 9th International Conference on Human Language Technologies – the Baltic Perspective, Numéro 22/09/2020, 2020
Éditeur: IOS Press

A Comparative Study of Speech Anonymization Metrics

Auteurs: Maouche, Mohamed; Srivastava, Brij Mohan Lal; Vauquier, Nathalie; Bellet, Aurélien; Tommasi, Marc; Vincent, Emmanuel
Publié dans: INTERSPEECH 2020, Numéro 25/10/2020, 2020
Éditeur: ISCA

On Semi-Supervised LF-MMI Training of Acoustic Models with Limited Data

Auteurs: Sheikh, Imran; Vincent, Emmanuel; Illina, Irina
Publié dans: INTERSPEECH 2020, Numéro 25/10/2020, 2020
Éditeur: ISCA

Achieving Multi-Accent ASR via Unsupervised Acoustic Model Adaptation

Auteurs: Turan, M. A. Tuğtekin; Vincent, Emmanuel; Jouvet, Denis
Publié dans: INTERSPEECH 2020, Numéro 25/10/2020, 2020
Éditeur: ISCA

Privacy Guarantees for De-identifying Text Transformations

Auteurs: Adelani, David Ifeoluwa; Davody, Ali; Kleinbauer, Thomas; Klakow, Dietrich
Publié dans: INTERSPEECH 2020, Numéro 25/10/2020, 2020
Éditeur: ISCA

Who Started this Rumor? Quantifying the Natural Differential Privacy Guarantees of Gossip Protocols

Auteurs: Bellet, Aurélien; Guerraoui, Rachid; Hendrikx, Hadrien
Publié dans: 34th International Symposium on Distributed Computing, Numéro 12/10/2020, 2020
Éditeur: LIPIcs

Fully Decentralized Joint Learning of Personalized Models and Collaboration Graphs

Auteurs: Zantedeschi, Valentina; Bellet, Aurélien; Tommasi, Marc
Publié dans: 23rd International Conference on Artificial Intelligence and Statistics, Numéro 26/08/2020, 2020
Éditeur: PMLR

Anonymisation and re-identification risk for voice data

Auteurs: Moretón, Alvaro; Jaramillo, Ariadna
Publié dans: European Data Protection Law Review, Numéro 19/07/2021, 2021, ISSN 2364-284X
Éditeur: Lexxion

MasakhaNER: Named Entity Recognition for African Languages

Auteurs: David Ifeoluwa Adelani; Jade Abbott; Graham Neubig; Daniel D’souza; Julia Kreutzer; Constantine Lignos; Chester Palen-Michel; Happy Buzaaba; Shruti Rijhwani; Sebastian Ruder; Stephen Mayhew; Israel Abebe Azime; Shamsuddeen H. Muhammad; Chris Chinenye Emezue; Joyce Nakatumba-Nabende; Perez Ogayo; Aremu Anuoluwapo; Catherine Gitau; Derguene Mbaye; Jesujoba Alabi; Seid Muhie Yimam; Tajuddeen Rabiu
Publié dans: Transactions of the ACL, Numéro 07/10/2021, 2021, ISSN 2307-387X
Éditeur: MIT Press
DOI: 10.1162/tacl_a_00416

How can Private Information Recorded by Voice-Enabled Systems be Identified?

Auteurs: Moretón Poch, Álvaro; Jaramillo, Ariadna
Publié dans: European Data Protection Law Review, Numéro to appear, 2020, ISSN 2364-2831
Éditeur: Lexxion

Monolingual and cross-lingual intent detection without training data in target languages

Auteurs: Jurgita Kapočiūtė-Dzikienė; Askars Salimbajevs; Raivis Skadiņš
Publié dans: Electronics, Numéro 11/06/2021, 2021, ISSN 2079-9292
Éditeur: MDPI
DOI: 10.3390/electronics10121412

The VoicePrivacy 2020 Challenge: Results and findings

Auteurs: Tomashenko, Natalia; Wang, Xin; Vincent, Emmanuel; Patino, Jose; Srivastava, Brij Mohan Lal; Noé, Paul-Gauthier; Nautsch, Andreas; Evans, Nicholas; Yamagishi, Junichi; O'Brien, Benjamin; Chanclu, Anaïs; Bonastre, Jean-François; Todisco, Massimiliano; Maouche, Mohamed
Publié dans: Computer Speech and Language, Numéro to appear, 2022, ISSN 1095-8363
Éditeur: Elsevier

Recherche de données OpenAIRE...

Livrables

Publications

Partager cette page

Télécharger