Oxford Wave Research

11/07/2025

𝗙𝗼𝗼𝗱 𝗳𝗼𝗿 𝘁𝗵𝗼𝘂𝗴𝗵𝘁 𝘀𝗼 𝘀𝘄𝗲𝗲𝘁 𝘆𝗼𝘂 𝘄𝗼𝗻’𝘁 𝗻𝗲𝗲𝗱 𝘁𝗵𝗲 𝗰𝗼𝗳𝗳𝗲𝗲!

We are excited to be heading to for the Annual Conference of the International Association for Forensic Phonetics and Acoustics (IAFPA) from 20th to 23rd July 2025.

This year, we’ll be serving up four talks and a poster on deepfakes, scalable speaker discovery, and automatic transcription evaluation.

See the full menu below:

"𝙁𝘼𝙐𝙓𝗗𝗜𝗢: 𝗔𝗻 𝗮𝘂𝗱𝗶𝗼 𝗱𝗲𝗲𝗽𝗳𝗮𝗸𝗲 𝗱𝗲𝘁𝗲𝗰𝘁𝗼𝗿 𝗳𝗼𝗿 𝗹𝗮𝘄 𝗲𝗻𝗳𝗼𝗿𝗰𝗲𝗺𝗲𝗻𝘁 𝗮𝗻𝗱 𝗳𝗼𝗿𝗲𝗻𝘀𝗶𝗰𝘀" | Presented by Anil Alexander, Linda Gerlach, and our team.

"𝗧𝗼𝘄𝗮𝗿𝗱𝘀 𝗮𝗻 𝗶𝗻𝘁𝗲𝗿𝗽𝗿𝗲𝘁𝗮𝘁𝗶𝗼𝗻 𝗳𝗿𝗮𝗺𝗲𝘄𝗼𝗿𝗸 𝗳𝗼𝗿 𝗳𝗼𝗿𝗲𝗻𝘀𝗶𝗰 𝗮𝘂𝗱𝗶𝗼 𝗱𝗲𝗲𝗽𝗳𝗮𝗸𝗲 𝗱𝗲𝘁𝗲𝗰𝘁𝗶𝗼𝗻" | Presented by Finnian Kelly, Anil Alexander, and our team.

"𝗗𝗶𝘀𝗰𝗼𝘃𝗲𝗿𝘆 𝗮𝗻𝗱 𝗿𝗲𝘁𝗿𝗶𝗲𝘃𝗮𝗹 𝗼𝗳 𝘀𝗽𝗲𝗮𝗸𝗲𝗿𝘀 𝗳𝗿𝗼𝗺 𝗹𝗮𝗿𝗴𝗲 𝘂𝗻𝗹𝗮𝗯𝗲𝗹𝗹𝗲𝗱 𝗱𝗮𝘁𝗮𝘀𝗲𝘁𝘀 𝘂𝘀𝗶𝗻𝗴 𝘀𝗰𝗮𝗹𝗮𝗯𝗹𝗲 𝗰𝗹𝘂𝘀𝘁𝗲𝗿𝗶𝗻𝗴" | Presented by Thomas Coy, Finnian Kelly & Anil Alexander.

"𝙏𝙝𝙖𝙣𝙠 𝙮𝙤𝙪 𝙛𝙤𝙧 𝙬𝙖𝙩𝙘𝙝𝙞𝙣𝙜: 𝗔𝘂𝘁𝗼𝗺𝗮𝘁𝗶𝗰𝗮𝗹𝗹𝘆 𝗲𝘃𝗮𝗹𝘂𝗮𝘁𝗶𝗻𝗴 𝘁𝗿𝗮𝗻𝘀𝗰𝗿𝗶𝗽𝘁𝗶𝗼𝗻𝘀 𝗳𝗼𝗿 𝗵𝗮𝗹𝗹𝘂𝗰𝗶𝗻𝗮𝘁𝗶𝗼𝗻𝘀 𝗮𝗻𝗱 𝗺𝗶𝘀𝘀𝗶𝗻𝗴 𝗺𝗲𝗮𝗻𝗶𝗻𝗴" | Presented by Jadd Virji & Finnian Kelly.

13/06/2025

Getting our ducks in a row to go to UK and Ireland Speech (UKIS) on Monday. Looking forward to some great talks and posters!

https://sites.google.com/york.ac.uk/ukis2025

University of York

Source image (modified): "York University New Part Ron Cooke Hub" by Peter Church is licensed under CC BY-SA 3.0 via Wikimedia Commons.

19/05/2025

"𝐏𝐞𝐫𝐜𝐞𝐩𝐭𝐮𝐚𝐥𝐥𝐲 𝐢𝐦𝐩𝐨𝐫𝐭𝐚𝐧𝐭 𝐕𝐎𝐂𝐀𝐋𝐈𝐒𝐄 𝐟𝐞𝐚𝐭𝐮𝐫𝐞𝐬 𝐬𝐡𝐨𝐰 𝐛𝐞𝐭𝐭𝐞𝐫 𝐬𝐩𝐞𝐚𝐤𝐞𝐫 𝐝𝐢𝐬𝐜𝐫𝐢𝐦𝐢𝐧𝐚𝐭𝐢𝐨𝐧 𝐨𝐟 𝐩𝐢𝐭𝐜𝐡-𝐦𝐚𝐧𝐢𝐩𝐮𝐥𝐚𝐭𝐞𝐝 𝐟𝐚𝐤𝐞 𝐬𝐩𝐞𝐞𝐜𝐡."

Just over a week ago, we had the pleasure of hosting PhD student Chloe Patman from the Phonetics Lab of the University of Cambridge at our office. Chloe is particularly interested in the acoustic and automatic analysis of female voices for forensic casework. She ran a variety of experiments with recordings that were both humanly and artificially manipulated to change their pitch using our VOCALISE forensic automatic speaker recognition software. She had some very interesting findings, and we look forward to hearing more.

14/03/2025

What an incredible few days at Security and Policing 2025!

Once again, it was an incredible experience to connect with industry leaders, government officials, and security professionals, all working toward a safer future. From insightful discussions to cutting-edge technology showcases, the event was both inspiring and impactful. A huge thank you to everyone who stopped by, shared knowledge, and made this event so valuable. Looking forward to continuing these conversations and building on the innovations we explored!

And a special thanks to the dedicated and hard-working organizers, ADS Group Ltd, UK Defence & Security Exports and Joint Security and Resilience Centre (JSaRC), for putting together such a well-run and meaningful event.

06/03/2025

Security and Policing 2025 is just around the corner!

This year, we are excited to showcase our cutting edge audio and speech processing tools:

TEAPOT: Our latest on-premise speech-to-text, transcription, translation and analysis software.

Fauxdio Detector: Audio deepfake detection software which produced near perfect accuracy on unseen challenge data in the audio modality at the recent Home Office Deepfake challenge. https://www.gov.uk/government/case-studies/innovating-to-detect-deepfakes-and-protect-the-public

VOCALISE, WHISPERS and PHONATE: Our well-known and well-loved forensic and investigative voice biometric tools.

SMARTSUITE: Powerful forensic audio enhancement suite.

24/02/2025

04/02/2025

🎉 Calling All Aspiring Scientists! 🌟

We’re excited to be a part of the VoCS (Voice Communication Sciences) project and are happy to announce that the applications are now open for 19 doctoral candidates!

Here’s your chance to join a cutting-edge international research initiative and elevate your career in the ground breaking field of Voice Sciences.

What is VoCS?

VoCS (Voice Communication Sciences) is a prestigious 4-year project funded by the European Union’s Horizon Europe Framework through the Marie Skłodowska-Curie Grant. This remarkable initiative unites 22 academic and non-academic partners from 12 countries to push the boundaries of cognitive neuroscience, acoustics, phonetics, and computing science.

What’s on Offer?

We are recruiting for 19 doctoral positions as part of our Training Network, offering you a hands-on opportunity to engage in innovative research and contribute to transformative advancements in the field of Voice Sciences.

Eligibility Criteria:

🎓 Education Level: A Master’s degree (or equivalent) in neuroscience, biology, computer science, engineering, phonetics, communication sciences, or related disciplines.
🔬 Research Experience: 1-4 years.

🗓️ Applications Open: 1 February 2025
⏳ Deadline: 15 March 2025

Ready to Shape the Future?

Learn more about the program and : https://www.vocs.eu.com/

Join us and become a part of pioneering research in Voice Sciences, working in an innovative and collaborative international environment! 🚀

16/01/2025

VOCALISE on the big screen at University of York for the 'Automatic Speaker Recognition: Getting Started for Research' course?? The training team are prepared for challenging days with some of the best and brightest!

24/12/2024

06/12/2024

We recently contributed to a report on deepfakes and election security by The Alan Turing Institute’s Centre for Emerging Technology and Security (CETaS), titled "AI-Enabled Influence Operations: Safeguarding Future Elections".

In the words of the lead author Sam Stockwell: “Oxford Wave Research played an important role in helping us shape the research findings, particularly with suggesting technical recommendations for combatting deepfake threats during elections.”

The report highlights the growing threat of AI-enabled influence operations in elections, focusing on findings from the 2024 US presidential election, during which several audio deepfakes were used to spread disinformation. The report concludes that "AI technology may not have fundamentally reshaped or disrupted the outcomes of key elections in 2024, but this is no cause for complacency". AI disinformation had no measurable impact on election outcomes in all monitored countries, including the United States, United Kingdom, European Union, Taiwan, and India, though it amplified false narratives, inflamed political debates and reinforced ideological divides among voters.

We are proud to have played a small part in framing the final recommendations of this pivotal report, which encourages global efforts to combat AI-enabled disinformation and protect the integrity of democratic elections worldwide.

This CETaS Research Report investigates AI-enhanced hostile influence operations during election cycles, focusing on the 2024 US presidential election and fu

08/11/2024

We are thrilled to welcome Jadd Virji to Oxford Wave Research as a Research and Software Engineer! Graduating with First Class Honours in Linguistics from the University of Cambridge and a strong focus on NLP, Jadd brings expertise in multilingual large language models. After a successful internship with us in the summer of 2023, he is now working full-time with our R&D and software teams on innovative projects in speech recognition, audio deepfake detection and explainability, as well as forensic analysis. Welcome back, Jadd!

16/10/2024

🔊 Join Us for "Automatic Speaker Recognition: Getting Started for Research" 🔊

📅 Date: Thursday 16th & Friday 17th January 2025
📍 Location: University of York (In-person)

Are you a research student interested in forensic speech science or automatic speaker recognition? Then this hands-on workshop hosted in collaboration with the University of York Forensic Speech Science Team is the perfect opportunity to learn more about how automatic speaker recognition systems work, explore their applications in forensic casework, and get practical experience using our software VOCALISE.

Over the course of two days, you will gain insights into pre-processing, feature extraction, speaker modelling, scoring, and calibration in a mix of theoretical and practical sessions. Additionally, you will be able to develop research questions and conduct tests relevant to your research.

🎓 Who is this for?
Current Masters or PhD students interested in automatic speaker recognition systems.

💻 Sign up now to secure your spot!
👉 https://docs.google.com/forms/d/e/1FAIpQLSe-NTCaRWA88cjd3UpxQgW6daqMC38VIpl_A0bi_H4sImA-Sw/viewform

ℹ️ For more information visit: https://www.york.ac.uk/business/cpd/sector-specific-courses/automatic-speaker-recognition/