Oxford Wave Research

13/09/2024

An ecstatic end to four years of hard work - Linda Gerlach celebrating after completing her PhD viva at the University of Cambridge University of Cambridge. She completes the Oxford Wave Research PhD studentship with Selwyn College Cambridge! Pictured with her supervisors Kirsty McDougall, Anil Alexander, and Finnian Kelly at the Cambridge Phonetics Lab and the beautiful lawns at Selwyn College Cambridge.

23/08/2024

Oxford Wave Research’s audio deepfake detection technology excelled in the Home Office-led ‘Deepfake Detection Challenge’. We topped the competition leader board for audio in the challenge determining which elements of a digital asset are deepfake. The challenge, open to UK companies, universities and researchers, required processing millions of deepfake and real images, audio and video files to build and test deepfake detection technology solutions. The UK Home Office, The Alan Turing Institute and the Department for Science, Innovation and Technology created this challenge and sought to seek out ‘the best of the best’ in innovative solutions to tackle the current and emerging threats presented by the increasing use of deepfakes.

We competed in the audio category and were thrilled with our near perfect performance. Along with five other organisations, we were invited to showcase and discuss our solution at the Deepfake Detection Challenge Showcase event at the iconic Ministry of Sound venue in London in July. Our R&D team has been working on voice cloning and audio for several years now, as we believe recent developments in voice cloning will fundamentally impact fraud, terrorism, sexual exploitation, and political disinformation. We were delighted, therefore, that our solution was able to perform so well with the challenge data.

We had the privilege of meeting Baroness Jones of Whitchurch (Parliamentary Under-Secretary of State at the Department for Science) with whom we had a very engaging discussion and demonstration of our audio deepfake detection technology. We were very pleased to listen to talks from Rupert Shute (Home Office, Deputy Chief Scientific Adviser), Talitha Rowland (Director, Security & Online Harms, Department for Science, Innovation and Technology), and Professor Jennifer Rubin (Home Office, Chief Scientific Adviser), which highlighted the problems and the need for solutions to combat these threats.

We would like to thank Andrew Tyeloo (Head of AI R&D Programme, Home Office) and the Accelerated Capability Environment (ACE) team, the Department for Science, Innovation and Technology, The Alan Turing Institute, and all those who played a part in putting together this fantastic challenge.

For a detailed account of the showcase and the ground-breaking solutions presented, read the full blog post https://ace.blog.gov.uk/2024/07/30/innovative-solutions-unveiled-at-the-deepfake-detection-challenge-showcase/ and the LinkedIn post https://www.linkedin.com/posts/accelerated-capability-environment-ace_building-a-safer-digital-future-deepfake-activity-7224309577146601472-FsqM/

06/08/2024

Today we were honoured to host the Deputy Chief Scientific Advisor to the Home Office, Rupert Shute, whom we recently met at the Home Office Deepfake Challenge. It was great to get ‘deep, technical and geeky’ to discuss all things voice biometrics, deepfake audio detection, speech processing and audio fingerprinting. We were also delighted to know he shared our belief in Likelihood Ratios and Bayesian reasoning.

01/08/2024

After an engaging workshop and panel discussion on voice deepfakes at the International Association for Forensic Phonetics and Acoustics (IAFPA) conference our CEO Dr Anil Alexander was quoted in Canadian media in articles in L'Actualite (https://lactualite.com/actualites/quand-la-phonetique-et-lacoustique-aident-a-resoudre-des-enquetes/ - in French) and other outlets. He spoke about the threat posed by deep fake voices. detection, about our approaches to tackling these threats.

MONTRÉAL — Partir d'un enregistrement audio pour dresser le portrait de suspects potentiels. Voilà un des défis que doivent relever certains linguistes dont le rôle est d'appuyer les forces de l'ordre dans leur enquête. Le principe est «de partir d'une voix dans l'univers, puis de raffiner l...

12/07/2024

We are already looking forward to lively discussions during Anil's and Finnian's Audio Deepfake Workshop and Panel Discussion at IAFPA 2024 organised by UQAM | Université du Québec à Montréal and the Royal Canadian Mounted Police, taking place in the Coeur des sciences in Montreal on July 28-31, 2024!

25/06/2024

We are dipping into the Audio Engineering Society International Conference on Audio Forensics in beautiful Colorado with two talks that will hopefully make a splash!

🎙️ We have two long format talks on there by our CEO Dr Anil Alexander - one on managing variable speech quality and dealing with multi-speaker recordings in forensic speaker comparisons, and the other on ‘Congruence plots’, a new visual approach for comparing speaker recognition system outputs.

https://aes2.org/event-extra/2024-aes-8th-international-conference-on-audio-forensics/

19/06/2024

Day 1 at Forensics Europe Expo (FEE) completed: great to meet our customers old, new and to be!

17/06/2024

🔍🌟 Newsflash: It is here! The Forensics Europe Expo 2024! 🌟🔍

Interested in voice biometrics, deepfake audio detection and audio processing technologies? We’ll be showcasing our state-of-the-art technologies at the premier event for forensic science and investigation in Europe, taking place in London on the 19th and 20th of June.

This event brings together global forensic science solution providers and gives visitors the opportunity to explore the latest advancements in the field.

Don't miss the chance to experience our powerful solutions firsthand!

📍 Visit us at stand FEE518

13/06/2024

Can sound-alikes actually help in forensics?

In a study led by Linda Gerlach, we explore how choosing similar sounding speakers in a relevant population for comparison could impact the strength of the evidence.

Linda will be in Quebec, Canada for Odyssey 2024, to present this paper co-authored by her, Finnian Kelly, Kirsty McDougall, and Anil Alexander from Oxford Wave Research and the University of Cambridge. This study reveals how different population selection strategies including one based on automatically assessed perceptual voice similarity can impact discrimination and calibration of a forensic automatic speaker recognition system.

📻Stay tuned for more updates on upcoming conferences. 🧠💡

08/03/2024

We’re showcasing our cutting-edge products in forensic voice biometrics, transcription and translation, audio fingerprinting and processing at Security & Policing, the official UK Home Office Event, between 12‐14 March 2024 in Farnborough. Visit us at stand P73

For more information: https://www.securityandpolicing.co.uk/

20/02/2024

SpectrumView, our hugely popular (1 million+ downloads) iOS audio analysis app for iPhone, iPad, and now even Vision Pro, has received a new update.

SpectrumView users have shared their experiences, from using it in the depths of the North Sea for underwater acoustics, to the forests of Vietnam for listening to gibbons, and even to the outskirts of the Montserrat Volcano for recording insect sounds!

Following a complete overhaul to ensure compatibility with the latest iOS versions and fix bugs, we've also included one of the most requested features: now, you can import and analyse audio from videos in your device library.

With SpectrumView, the only limit to its potential uses is your imagination!

02/02/2024

🎙️Tune in! Oxford Wave Research on the airwaves with WRFM- Radio for Oxfordshire & The Cotswolds.

We are thrilled to have been featured on WRFM’s “Science on the Radio”! 📻 A big shoutout to hosts Nina and Grant for the engaging interview with our CEO, Dr Anil Alexander, exploring the distinction between speech recognition (understanding what is being said) and speaker recognition (identifying who is speaking), technical aspects of voice analysis, application of voice recognition in sectors like banking and forensics, and the importance of combining human expertise with machine learning to achieve the best results.

Curious about the science behind speaker and speech recognition? Listen in to the podcast here 🎧:

https://www.mixcloud.com/witneyradio/wrfm-science-on-the-radio-episode-42-oxford-wave-research/

Most frequently used words in the interview in the picture below (generated by our speech product TEAPOT below)

02/02/2024

Ever wondered whether AI can really copy your voice, how do law enforcement agencies use voice data in trials, are humans needed anymore in audio? FInd out more on tomorrow's Science on the Radio at 10am. We chat to the clever people at Oxford Wave Research

19/12/2023

Thank you for your continuous trust and support throughout 2023, we wish you all a very happy and peaceful festive break and look forward to working with you in 2024.

11/12/2023

Our principal research scientist, Dr Finnian Kelly, has returned (sushi) rolling in new ideas from the Joint Workshop of VoicePersonae and ASVspoof held in Tokyo last week!

This event brought together researchers from both speech synthesis and deepfake detection communities, which prompted many interesting discussions about the future of synthetic speech and the implications for speech technology.

Some of the latest advancements in speech synthesis were presented, along with some really positive example use cases like voice banking. At the same time, there were many promising directions for the detection of synthesised speech (aka deepfake detection) proposed, including new neural architectures, watermarking, and explainability approaches to involve a human in the loop.

We'd like to convey a big thank you to Junichi Yamagishi and the team at NII for hosting this excellent event, and we look forward to bringing some new ideas to our work in this space!

国立情報学研究所 National Institute of Informatics

30/11/2023

Our CEO, Dr Anil Alexander had the privilege of speaking at The Royal Society, one of the world’s most prestigious and oldest scientific academies, on the topic of voice recognition in forensic science last month and the recordings for the talks are now available on the The Royal Society’s Youtube channel (https://youtu.be/pXdPoCcFlrw?si=h9Pgwh4g9t6tkXIN&t=3514). He was speaking at the "Science in the Interests of Justice" conference, co-organised by the Royal Society and the National Academy of Sciences as part of their Science and Law programme which brings together top scientists and legal experts from both sides of the Atlantic to explore the crucial role of science in the judicial system.

08/11/2023

Our CEO, Dr Anil Alexander was interviewed by BBC Radio 4's Joshua Rozenberg KC for the 'Law in Action' programme and spoke about forensic speaker recognition and fake voices. On this episode titled 'Deepfakes and the law', Joshua Rozenberg, Britains best-known commentator on the law, and listened to samples of mock-suspect and fake recordings played by Anil and tried to guess them right.

This episode airs live on BBC 4 on Thurs 9th Nov '23 at 20.00 GMT or you can listen to this episode right now on
BBC Sounds (https://www.bbc.co.uk/programmes/m001s5l9)
Spotify (https://open.spotify.com/episode/4w7D2oeRRexLVDbGO3Vjhk)
or Apple Podcasts (https://podcasts.apple.com/gb/podcast/deepfakes-and-the-law/id265307843?i=1000634023260)

Picture: Anil Alexander with Joshua Rozenberg (Presenter) and Arlene Gregorius (Producer) at the BBC's Broadcasting House in London.

07/11/2023

Sushi and Fake Speech: Unmasking the rolls of deception!

At OWR we are actively engaged in the problem of deepfake detection and speech synthesis, and have been working with academic and law enforcement partners on this for a few years now.

Our principal research scientist, Dr Finnian Kelly, will be presenting at the upcoming Joint Workshop of VoicePersonae and ASVspoof 2023 in Tokyo, where he will discuss some of the challenges and opportunities synthetic speech poses for forensics.

The workshop will bring together many of the leading researchers in deepfake detection and speech synthesis in a very interesting programme of talks, and we are looking forward to being a part of it!

The joint workshop of project VoicePersonae and ASVspoof challenge series, held in Hitotsubashi, Tokyo, Japan.

27/09/2023

Our CEO, Dr Anil Alexander is speaking at The Royal Society, one of the world’s most prestigious and oldest scientific academies, on the topic of voice recognition in forensic science. Dr Anil Alexander will be one of the invited speakers at the "Science in the Interests of Justice" conference, taking place on October 3-4, 2023. This is co-organised by the Royal Society and the National Academy of Sciences as part of their Science and Law programmes and brings together top scientists and legal experts from both sides of the Atlantic to explore the crucial role of science in the judicial system.

If you would like to be virtually part of this event, you can still secure a spot and attend online using the link below:

Science in the interests of justice - Science and the law meeting

22/09/2023

Where did I hear that voice before? The OWR team is spending an illuminating day learning about new research in earwitness evidence and voice parades at the Improving Voice Identification Procedures (IVIP) project knowledge exchange event organised by Selwyn College Cambridge University.

23/08/2023

...in the meantime at INTERSPEECH2023 in Dublin: all work and no play!

11/08/2023

INTERSPEECH 2023 is just around the corner and we’re excited to have two OWR co-authored papers with distinguished researchers from University of Cambridge and York at this year’s event in Dublin-

Linda Gerlach, Kirsty McDougall, Finnian Kelly, and Anil Alexander. Voice twins: discovering extremely similar-sounding, unrelated speakers

Vincent Hughes, Jessica Wormald, Paul Foulkes, Philip Harrison, Finnian Kelly, David van der Vloed, Poppy Welch, and Chenzi Xu. Automatic speaker recognition with variation across vocal conditions: a controlled experiment with implications for forensics

Our team will also be on hand to discuss the latest advancements in voice biometrics and audio processing technology. So please visit us on stand A1.

We look forward to seeing you there!

12/07/2023

Congratulations Linda Gerlach for winning the Oxford Wave Research prize for best student paper in acoustics and ASR at IAFPA 2023, and Daniel Lee who is with our OWR-Cambridge fellowship at Cambridge for winning the best student oral presentation.

Selwyn College Cambridge

07/07/2023

Oxford Wave Research staff are very excited to be attending the upcoming International Association of Forensic Phonetics and Acoustics IAFPA Conference, organised this year by the Centre for Forensic Phonetics and Acoustics (CFPA) at the University of Zurich and the Forensic Science Institute (FOR). We will be presenting a number of papers on our most recent work in the field of voice biometrics and audio processing. Have a look at the list of the presentations co-authored by the OWR researchers in collaboration with distinguished academicians and forensic scientists below:

"A convincing voice clone? Automatic voice similarity assessment for synthetic speech samples"
Linda Gerlach, Finnian Kelly, Kirsty McDougall, and Anil Alexander

"PASS (Phonetic Assessment of Spoofed Speech): Towards a human-expert-based framework for spoofed speech detection"
Daniel Denian Lee, Kirsty McDougall, Finnian Kelly, and Anil Alexander

"CON(gruence)-plots for assessing agreement between voice comparison systems"
Michael Jessen, Anil Alexander, Thomas Coy, Oscar Forth, and Finnian Kelly

"Automatic Speaker Recognition: does dialect switching matter?"
Marlon Siewert, Linda Gerlach, Anil Alexander, Gea de Jong-Lendle, Alfred Lameli, and Roland Kehrein

"Impact of the mismatches in long-term acoustic features upon different-speaker ASR scores"
Chenzi XU, Paul Foulkes, Philip Harrison, Vincent Hughes, Poppy Welch, Jessica Wormwald, Finnian Kelly, and David van der Vloed

"Effects of vocal variation on the output of an automatic speaker recognition system"
Vincent Hughes, Jessica Wormwald, Paul Foulkes, Philip Harrison, Poppy Welch, Chenzi Xu, Finnian Kelly, and David van der Vloed

13/06/2023

Looking forward to meeting the best speech brains in the UK and Ireland at UK Speech 2023 at The University of Sheffield this week, and happy to have been invited to speak about what we do in forensic and law enforcement audio.