Listen to Emmanuel Vincent, the project coordinator of COMPRISE, Nathalie Vauquier, the engineer in charge of Inria’s software development in COMPRISE, and Brij Srivastava, a young researcher who has joined Inria Startup Studio on October 1, 2021, to create a startup exploiting the results of the project. https://www.inria.fr/en/podcast-comprise-privacy-friendly-inclusive-voice-interface
The joint efforts of the COMPRISE consortium continue bearing results as the COMPRISE solutions are used in an E-Commerce app created by our partner Netfective Technology. This app demonstrates for companies that wish to expand their customer base beyond a single language the possibilities of adding privacy aware multilingual voice…
The joint efforts of the COMPRISE consortium continue bearing results as the COMPRISE solutions are used in E-Health apps created by our partner TILDE. These two apps demonstrate the possibility for companies that support medical institutions to add privacy aware multilingual voice enabled features cost efficiently using the COMPRISE solution….
The joint efforts of the COMPRISE consortium continue bearing results as the COMPRISE are used in consumer apps created by our partner Ascora GmbH. These three apps demonstrate the possibility for software developing companies and freelance developers to add privacy aware multilingual voice enabled features cost efficiently using the COMPRISE…
The sixth COMPRISE plenary meeting was held on October 21, 2021 remotely and on location at Saarbrücken. We were thrilled to welcome seven Advisory Board members from ELDA, CrossLang, DATAMED, FVLLMONTI, /e/OS, CNIL & CHRU-Lille.
Dietrich Klakow from Saarland University has been invited to participate at the EBDVF 2021 session on “Data and AI to remove language barriers and address societal challenges” ! EBDVF 2021 will be an excellent opportunity to promote COMPRISE to the broader AI and data community.
COMPRISE has just published an article titled “Data labelling: the hidden cost of voice-enabled technologies” in Technology.org This article talks about the methods used, so that a computer can understand speech. Click here to read the article!
How Technology is Changing Court Reporting A Brief history of Court Reporting and Stenography As Adam Brooks writes in his essay The History of Court Reporting and Stenography, technological advancements in the world of court reporting and stenography begin almost with the beginning of writing itself. In ancient Mesopotamia, the…
We have the pleasure to announce the acceptance of our Dagstuhl Seminar proposal entitled “Privacy in Speech and Language Technology”! The Seminar, which was initially scheduled on January 31 – February 5, 2021, will eventually take place on August 21-26, 2022. It will cover the fields of computer science and the…
Cloze-style Zero-shot Named Entity Recognition “You Can Call Me Al” is the title of a 1980’s hit pop song by Paul Simon, and it’s not an unusual phrase to say. Unlike most things, people normally have names. That’s why, if someone wishes to write something down, you’re much more likely…
COMPRISE has just published an article titled “Anonymisation and Re-Identification Risk for Voice Data” in Volume 7 (2021), Issue 2 of the European Data Protection Law Review (EDPL). This article talks in-depth about the challenges of anonymising voice data in relation to the GDPR. It focuses on the automatic anonymisation…
COMPRISE has just published an article titled “Enabling Voice-Based Apps with European Values” in the 126th issue of the ERCIM News magazine. This article gives a brief introduction to the challenges, and our solutions, in preserving user privacy while maintaining the usability of dialogue data. Click here to read the…
The fifth COMPRISE plenary meeting was held remotely on June 28, 2021. We were thrilled to welcome the following advisory board members: Gaël Duval (e.foundation) Johannes Fischer (Fraunhofer IIS, SPEAKER Project) Tom Vanallemeersch (ELRC Project) Georgeta Bordea and Jean-Luc Rouas (FVLLMONTI Project)
The joint efforts of the COMPRISE consortium continue bearing results as the COMPRISE Speech-to-Text Translation has been released by our partner TILDE. The COMPRISE Speech-to-Text Translation combines Speech-to-Text and Machine Translation in a smart manner in order to allow a user to speak his/her own language when interacting with a…
Protecting User Privacy with Voice Conversion With the increasing popularity of smart devices, more and more users have access to voice-based interfaces. Voice-based interfaces offer simple access to modern technologies and enable the development of new services. The building blocks behind these speech-based technologies are no longer handcrafted but learned…
Data annotation for Speech-to-Text and Natural Language Understanding What is Machine Learning? Machine learning refers to “Training AI algorithms to make decisions or predictions based on large amounts of data”. There are several types of machine learning techniques, depending on the amount of human knowledge provided as input: supervised learning,…
The COMPRISE Speech-to-Text-Translation: Discover the benefits! What is it? COMPRISE Speech-to-Text-Translation combines “Speech-to-Text” and “Machine Translation” in a smart way, to allow a user to speak his or her own language when interacting with a dialogue system that is internally designed for a different language. Why do I need it?…
Ethical concerns related to the use of voice technologies The ethical use of voice technologies, such as speech and voice recognition, is becoming more important every day. Devices such as smart speakers, smartphones or smartwatches collect massive amounts of data from users thanks to the wide range of activities they…
COMPRISE has provided input to the European Data Protection Board’s (EDPB) consultation on its draft guidelines on virtual assistants (“Guidelines 02/2021 on Virtual Voice Assistants”) published on 9th March, 2021. Check at the following link the comments provided by COMPRISE! https://edpb.europa.eu/sites/default/files/webform/public_consultation_reply/edpb_feedback_by_comprise.pdf
Today, Emmanuel Vincent introduced COMPRISE and the impact of European funding on research and innovation to a large audience of academic and industry researchers. This was part of a series of webinars on Horizon Europe organized by Region Grand Est. Today’s webinar called “Investing in digital technologies: from research to production, for…
COMPRISE has just published an article titled “The accent gap problem in minorities and dialect speakers” in Technology.org Click here to read the article!
Recent times have seen a remarkable growth of interest for smart software that offers advanced features built upon cutting-edge artificial intelligence (AI) and machine learning (ML) methods, such as the ability to detect and adapt to voice commands, parse natural language, classify video and images, forecast more accurately complex financial…
The joint efforts of the COMPRISE consortium continue bearing results as the COMPRISE Weakly Supervised Speech-to-Text (STT) and COMPRISE Weakly Supervised Natural Language Understanding (NLU) Tools have been released have been released. The COMPRISE Weakly Supervised STT and COMPRISE Weakly Supervised NLU are innovative software tools for automatic data transcription…
The Accent Gap: What it is, and how to tackle it. Here is a quick introduction to the Accent Gap. What is the Accent Gap? The accent gap is the inability of voice technologies to understand speakers with non-native or regional accents with the same accuracy as the majority of…
Overview The COMPRISE Cloud Platform provides a highly scalable, open-source, cloud-based solution for the management of multilingual speech and text data and models. The COMPRISE Cloud Platform offers a web application programming interface (API) and a web-based user interface (UI) which allow users to upload, store and manage speech and…
Today, Emmanuel Vincent introduced COMPRISE to 8th grade pupils at Saint-Léon school in Nancy, France. This was an opportunity to tell pupils about the inner working and the limitations of current voice technologies and to explain them the role of research and innovation in building the future of Europe.
The joint efforts of the COMPRISE consortium continue bearing results as the COMPRISE Cloud Platform has been released. The COMPRISE Cloud Platform, developers can manage multilingual speech and text data and models to create multilingual, voice-enabled applications in a faster, cost-effective, privacy-driven way. Check our press-release about our released tool!
Emmanuel Vincent from INRIA gave a great presentation about COMPRISE at a seminar series at the Japanese National Institute of Advanced Industrial Science and Technology (AIST)
Voice-based applications are still important. In fact, a survey by Adobe [Ref1] shows that during the COVID-19 pandemic the importance of voice-based applications has increased, thanks to their contactless interaction method and the desire for a safer, sanitary alternative. As much as one in three voice technology users mentioned hygiene…
The fourth COMPRISE plenary meeting was held remotely on January 12, 2021. We were thrilled to welcome the following advisory board members: Paul Andrey (CHU de Lille) Victoria Arranz (ELDA/ELRA) Kaspars Grosu (DATAMED)
The concept of data anonymisation Broadly speaking, data anonymisation can be defined as the process of irreversibly removing identifiable (confidential or private) information of both natural and legal persons[1] from raw datasets (e.g., text databases, audio recordings, etc.) [Ref1]. But there is more about anonymisation than what meets the eye,…
Discover COMPRISE Weakly Supervised NLU, an innovative automatic data labelling and model training software for Natural Language Understanding (NLU). Today’s Deep Learning technology is hungry for data – the more, the better. But for many tasks, raw data is not enough: every sample of data needs to be labelled first,…
The Cost Factor behind Speech-to-Text Technology Voice assistants equipped with Speech-to-Text technology have seen a major boost in performance and usage, thanks to the new powerful machine learning methods based on deep neural networks. These methods follow a supervised learning approach, requiring large amounts of paired speech-text data to train…
Emmanuel Vincent (COMPRISE’s Project Coordinator) has just published an article titled “”Peut-on faire confiance aux IA ?” (the translation in English is “Can AI systems be trusted?”) in The Conversation. Read it at the following link!
COMPRISE SDK Thanks to COMPRISE SDK, developers can create multilingual, voice-enabled applications in a faster, cost-effective, and privacy-driven way. The SDK is made for Smartphone applications developed with the Ionic framework, with Angular as a foundation. It consists of: the COMPRISE Personal Server, which allows the execution of large related…
The emergence of e-commerce platforms combined with the mobile revolution has changed the way millions of users around the world discover, select and order goods. These platforms are evolving at an incredibly fast pace to become a complete environment in which end-users (i.e., customers) are able to place orders within…
Voice technologies are being used in fields as diverse as medicine and education, but also for pure leisure. This article offers an overview of how categorisation and contextualisation can be used as methods not only to identify personal information but to design private-by-design voice-based solutions that intend to neutralise personal data and information/words that…
The joint efforts of the COMPRISE consortium continue bearing results as COMPRISE SDK has been released. With COMPRISE SDK, developers can create multilingual, voice-enabled applications in a faster, cost-effective, privacy-driven way. Check our press-release about our released tool!
In a world with more than 7,000 languages, roughly half of the people speak only one language, thus creating huge barriers that restrict cross-cultural communication. Thanks to advances in Artificial Intelligence and Machine Learning, we are steadily moving towards a future where hosting a dinner party for 20 guests, each…
Not just yet another note taking app Note taking apps are very useful in many situations. They can be used as a creative backlog for the user’s job, as a to-do list for household tasks or maybe as a shopping list. There are as many options as the number of…
The Call For Papers for the Privacy Preserving Machine Learning Workshop at NeurIPS 2020 (co- organized by our partner Aurelien Bellet, from INRIA) is open now !! This one day workshop focuses on privacy preserving techniques for machine learning and disclosure in large scale data analysis, both in the distributed…
Following the release of their white paper on voice assistants, the French data protection authority CNIL and Le VoiceLab will host a meetup on September 7 at 18:00 addressing the ethical, technical, and legal issues raised by voice assistants. Emmanuel Vincent, whose interview can be read in the white paper, will present ongoing…
Overview Are you a developer? Thanks to the COMPRISE SDK, you can now create multilingual, voice-enabled applications in a faster, cost-effective, and privacy-driven way. The SDK is made for Smartphone applications developed with the Ionic framework, with Angular as a foundation. It consists of: The COMPRISE Client Library, which can…
Six papers showcasing COMPRISE’s research advances have been accepted for publication at Interspeech 2020 ! Privacy guarantees for de-identifying text transformations [David Adelani, Ali Davody, Thomas Kleinbauer, Dietrich Klakow] A comparative study of speech anonymization metrics [Mohamed Maouche, Brij Mohan Lal Srivastava, Nathalie Vauquier, Aurélien Bellet, Marc Tommasi, Emmanuel Vincent] On semi-supervised…
The COMPRISE Voice Transformer and the COMPRISE Text Transformer protect both the voice of the users and their personal information. These software tools can be easily integrated into existing voice technologies, and provide developers with open source, validated, private-by-design tools. The development of privacy-enhancing technologies in Europe is a crucial step forward in ensuring the resilience…
Voice-enabled technologies have become incredibly popular in the last few years. They are present in a wide variety of devices such as smartphones, house appliances, or smart speakers, and commonly in the form of voice assistants. Voice-enabled technologies can truly facilitate everyday tasks such as checking up the weather or…
Additional papers for the COMPRISE project have been accepted! The most recent ones will be published at the 23rd International Conference on Text, Speech and Dialogue (TSD’2020), and are the following: “Investigating the Impact of Pre-trained Word Embeddings on Memorization in Neural Networks” [Aleena Thomas, David Adelani, Ali Davody, Aditya Mogadala,…
Main Problems in Multi-Accent Speech-to-Text Systems English is one of the most widely spoken languages worldwide and there exists a large population of non-native speakers in English compared to other languages. However, accented speech samples may have heterogeneous characteristics due to the mother tongue or the proficiency of the speakers….
The third COMPRISE plenary meeting was held remotely on June 03-04, 2020. We were thrilled to welcome the following advisory board members: Georg Rehm (META-NET) Fredrik Kronlid and Staffan Larsson (Talkamatic) Norbert Pfleger (Paragon Semvox) Eren Gölge (Mozilla)
Introduction Voice-based applications are creating massive growth opportunities for European SMEs by enabling natural interaction with an ever-increasing range of services. Thanks to its naturalness and efficiency, voice has become a common means of accessing a range of services such as web search, car navigation, and smart home control. This…
We are co-organizing a special issue on Voice Privacy that solicits papers describing advances in privacy protection for speech processing systems, including theoretical developments, algorithms or systems. Examples of topics relevant to the special issue include (but are not limited to): formal models of speech privacy preservation, privacy-preserving speech feature extraction,…
How we did and what we learned from building a Covid-19 chatbot When hit by an unexpected crisis like the global spread of a new infectious disease, people are in dire need of trustworthy and reliable information, especially since there is no scenario or action plan ready to tackle the…
> The operating branch runs a spoken dialogue system on the user’s device or on a personal server. The dialogue outcomes are sent to the company delivering the desired service only. The training branch removes personal information from voice data using privacy driven text and voice transformers. The transformed data are sent to the COMPRISE Cloud-based Platform to train the spoken dialogue system. COMPRISE VOICE TRANSFORMER Prevents…
Since the invention of the very first smartphone, many things have changed. While the first apps were only provided by the developers themselves, there are nowadays app stores with millions of apps for nearly everything [Ref1]. Challenges when developing consumer applications While developing an app, the developer has to think…
In our previous posts [Post1] and [Post2], we have presented a rather streamlined version of the issue at hand and introduced quite a few simplifications along the way. Still, we have seen that privacy transformations of text are complex. A number of challenges presented themselves and we have discussed a…
Registration is still open to take part in the first international challenge on speech data privacy. The VoicePrivacy initiative is spearheading the effort to develop privacy preservation solutions for speech technology. It aims to gather a new community to define the most effective processes and metrics, while also benchmarking the existing privacy-enabling solutions and creating…
Removing sensitive portions from a text, sometimes known as “sanitizing”, is often done by simply blackening the relevant words. Many people have seen such a redacted document at some point, it’s a technique commonly applied to classified government reports, for instance. A prominent example of such a document from recent…
“Tell me what you read and I tell you who you are.” – Pierre de La Gorce Voice assistants, such as Alexa, Siri, or Google Assistant, are becoming increasingly popular. Some users are, however, worried about the fact that their vocal interactions with these devices often get stored in the…
Today, voice-enabled technologies rely on a machine learning paradigm called deep learning. Deep learning has led to major improvements in speech-to-text, spoken language understanding, and dialogue management. Before moving into the GDPR principles, first, let’s first briefly recall the concepts of machine learning and deep learning introduced and detailed in…
Compatibility and integration are becoming essential features for voice technologies. The expected increase in the development of mid-level devices, which are able to connect with smart speakers, although not being fully smart themselves, serves as the perfect example [Ref1]. Despite the popularity of voice technologies, 41% of voice users report concerns on trust, privacy and…
The second COMPRISE plenary meeting was held on December 05-06, 2019 in Bordeaux. We welcomed three advisory board members who are Frédéric Peak (DYDU), Mario Fritz (CISPA) and Kaspars Grosu (Children’s Clinical University Hospital, Riga).
Speech representation In order to process speech on computers, the microphone output is collected and represented as a waveform, which is nothing but a sequence of floating point values called amplitudes. Here is a snapshot of the Praat software which is commonly used for speech analysis. The top image shows…
Emmanuel Vincent from INRIA gave a great presentation, at VoiceTech, about the major scientific and technological challenges in speech processing that we are facing today. COMPRISE as an initiative done by our Consortium members was presented in this context!
Emmanuel Vincent from INRIA was invited for a debate entitled “What voice assistants really know about us?”. Emmanuel explained the idea behind the COMPRISE project by highlighting the need to create a methodology that will protect the users’ data, in order to ensure their privacy. This was broadcast via the Rue89 Strasbourg media! Check…
To perform the tasks requested by users in the digital economy, voice-based technologies record and process large amounts of data, including personal data, which is centrally stored in the cloud. Voice-enabled systems capture the user’s speech signal (voice), but also background sounds! Background sounds can provide private information regarding the user…
Healthcare has been one of the countless beneficiaries of the revolutionary advances that widespread computing has brought. Fast, efficient data organisation, storage and access that have greatly sped up the medical enterprise, yet many low hanging fruits remain hanging. Chief among those is the increased application of technologies that can…
Our partner Tilde has organized a workshop for Riga’s Children’s Hospital. The workshop covered two main topics: (1) eventual use-case scenarios for two Tilde/COMPRISE demonstrators and (2) availability of data for system training and testing. Participants got brief introduction of the COMPRISE project and its objectives as well as the underlying technologies of…
Voice-based technologies are capable not only of executing and interpreting commands but also of identifying individuals by using their voice only [Ref 1]. Voice recognition (a.k.a. speaker recognition) should not be confused with speech recognition. The former aims to verify the identity of the person speaking while the latter aims…
Emmanuel Vincent from INRIA has been invited to participate at the Global Forum on AI for Humanity (GFAIH)! Experts from AI, social sciences, humanities, and engineering, as well as innovators, economic actors, policy makers and civil society representatives were all at GFAIH in order (i) to reach a common understanding of the perspectives…
As part of COMPRISE, the work “Private Protocols for U-Statistics in the Local Model and Beyond” [James Bell, Aurélien Bellet, Adria Gascon, Tejas Kulkarni] has been accepted for presentation at two workshops part of CCS’2019: Theory and Practice of Differential Privacy (TPDP) and Privacy Preserving Machine Learning (PPML). Congratulations to our team!…
Talking about dialogue systems such as Alexa, Siri, or Google Assistant raises several questions. In the recent years, these virtual personal assistants have become the most popular representatives of their field. Although the quality of the communication and the supported languages are improving day by day, there are still several…
Emmanuel Vincent, Irina Illina, and Brij Mohan Lal Srivastava demonstrated the privacy-driven voice transformation technology developed in COMPRISE as part of the Science festival at Université de Lorraine! It was a great interaction between INRIA’s members and several groups of junior high school students, students and staff of Université de Lorraine…
A great COMPRISE pitch is given @META-FORUM’2019 along with a booth! A brief overview of COMPRISE, its main goals, and fields of possible cooperation with the European Language Grid (ELG) platform were presented!
Our partner Dietrich Klakow from USAAR, expert in the field of language technologies, has been invited to participate at the EC expert consultation on language technologies and web accessibility! He gave an elevator pitch highlighting what can be realized within the next 10 or 20 years in the domain of language technology.
Ethics in retail Our partner Intermarché is one of the biggest food retailers in France, Belgium, Poland and Portugal. Intermarché is an ethics-centered retailer that aims to promote ethical values at all levels of the company from the production of goods to the customers’ basket through the whole supply chain…
On 25 May 2018, the General Data Protection Regulation (GDPR) came into force across the European Union. The new Regulation provides a broader definition of personal data with respect to its predecessor, the Data Protection Directive (DPD), and has rapidly become a standard for the processing of personal data not…
In July 2019, the Flemish news outlet VRT published an article revealing that internet giant Google employs thousands of people around the globe to listen to recordings of the customers – mostly unbeknownst to them. This came as quite a shock given that an almost identical report was previously published by the…
Voice assistants like Alexa, Siri, and Google Assistant are still a relatively new technology for the consumer market. However, under the label of “Spoken Dialog Systems” (SDS), it’s a research area that has quite a long history. The classical architecture of an SDS typically looks something like this: The user…
The first COMPRISE paper entitled “Privacy-Preserving Adversarial Representation Learning in ASR: Reality or Illusion?” [Brij Mohan Lal Srivastava, Aurélien Bellet, Marc Tommasi, Emmanuel Vincent] has been accepted for publication at Interspeech’2019! Great team work!
COMPRISE was broadcast on France 3 TV Channel! Emmanuel Vincent explained the aim of the project by highlighting the need to create a methodology that will protect the users’ data, in order to ensure their privacy.
The first COMPRISE plenary meeting was held on June 11-12, 2019 in Madrid. We welcomed three advisory board members who are Frédéric Peak (DYDU), Alice Coucke (Snips), and Félicien Vallet (CNIL). Plenary Meeting Agenda
COMPRISE was presented at the “Showcasing Language Technology in H2020 and CEF Telecom Projects” workshop that was held on the 26 of April 2019 in Brussels. The workshop introduced Digital Service Infrastructures and their stakeholders at Member State level to language technology related projects in the H2020 program and in CEF…
Marc Tommasi and Aurélien Bellet from INRIA are organizing the 10th edition of the privacy and data protection workshop. The workshop will take place from 9 to 11 July 2019 at Cape Hornu, in the heart of the Baie de Somme, in the Picardie region of France. The workshop is multi-disciplinary, and aims to bring…
Emmanuel Vincent gave a talk entitled Speech & deep learning : successes and grand challenges at the AI and Language for Citizens Day organized by Lorraine Université d’Excellence on March 15, 2019. Two of the four challenges listed in this talk, namely cost reduction and increased privacy, are core challenges of…
COMPRISE had its kick-off meeting on January 24-25. The meeting was hosted by Inria in Paris. Agenda Press Release
Emmanuel Vincent attended the networking session entitled Where Multilingualism, Big Data and Artificial Intelligence meet: Language Technologies for the Next Generation Internet organized by META-NET and LT Innovate at the ICT 2018 conference on December 4, 2018. A summary of the 7 projects accepted under the ICT-29-2018 call was presented, including COMPRISE. He met…
