1. Voice Biometry standard proposal Honza Černocký Brno University of Technology, BUT Speech@FIT, Czech Republic Sep 8 th 2015, Interspeech VBS meeting

2. Program Honza Cernocky – intro, “why?” Ondrej Glembek – Technical description Petr Schwarz – Phonexia remarks Discussion Honza Cernocky – next steps End 16.00, no buffet, drinks, entertainment  BUT Speech@FIT Honza Cernocky 05/20152/56

3. Situation In the last 10 years, scientific advances in speaker recognition (JFA, iVectors, PLDA) allowed for producing precise and robust SRE systems Quickly adopted by vendors, producing solutions that are successful on the market. R&D never stopping Everyone continuously improving performance of their system, robustness, calibration, etc New versions of engines released A vibrant community working in cooperative/competitive mode both for R&D labs and vendors. BUT Speech@FIT Honza Cernocky 05/20153/56

4. It works BUT Speech@FIT Honza Cernocky 05/20154/56 SID Score, hard decision … PepiVec Score, hard decision … CocaiVec SID

5. It does not work  BUT Speech@FIT Honza Cernocky 05/20155/56 SID PepiVec CocaiVec SID MISMATCH

6. Making it work BUT Speech@FIT Honza Cernocky 05/20156/56 CocaiVec SID Score, hard decision … CocaiVec

7. Making it really work – standardized iVectors BUT Speech@FIT Honza Cernocky 05/20157/56 SID VBSiVec SID Score, hard decision …

8. Making it really work – standardized iVectors BUT Speech@FIT Honza Cernocky 05/20158/56 SID VBSiVec SID Score, hard decision …

9. Making it really work – standardized iVectors BUT Speech@FIT Honza Cernocky 05/20159/56 SID VBSiVec SID Score, hard decision …

10. The main thing BUT Speech@FIT Honza Cernocky 05/201510/56 I-VECTOR EXTRACTION (VENDOR 1) COMPARISON AUDIO 1 I-VECTOR EXTRACTION (VENDOR 2) AUDIO 2 SCORE SPEAKER IDENTITY NO CONTENT SPEAKER IDENTITY AND CONTENT i-vector

11. What is needed Fix the core iVector extraction algorithms Fix the necessary parameters Do the necessary minimum, let people freedom to use their (own, best) VAD and scoring. Do it well for the core condition – telephone, not trying to address everything. BUT Speech@FIT Honza Cernocky 05/201511/56

12. We WANT Users Having interoperable systems Being able to exchange speaker information without compromising content within companies/agencies, across companies/agencies and across borders Vendors Increasing the whole market (think about introduction of USB!) R&D labs sharing iVectors between labs without lengthy discussions on configuration (not excluded though!) Giving a working recipe to juniors to play with. Obtaining massive data from the users BUT Speech@FIT Honza Cernocky 05/201512/56

13. We DON’T WANT stop R&D (both academic and commercial) of speaker recognition technology by saying that this will be the only iVector extraction scheme forever. all of us are trying to push the field further, sometimes as collaborators, sometimes as competitors. We want to define a snap-shot of the best practice up to day on which we could agree. Earn money on licenses or patents – the proposed standard is license and patent-free Have something too complex and too relying on a proprietary and/or 3 rd party technology. Present this as an ultimate forensic solution. BUT Speech@FIT Honza Cernocky 05/201513/56

14. What is there http://voicebiometry.org/ - technical description, Python code with all necessary parameters (feature extraction, UBM, T-matrix)http://voicebiometry.org/ Google group http://groups.google.com/d/forum/voice- biometry-standard - please subscribehttp://groups.google.com/d/forum/voice- biometry-standard BUT Speech@FIT Honza Cernocky 05/201514/56

15. Program Honza Cernocky – intro, “why?” Ondrej Glembek – Technical description Petr Schwarz – Phonexia remarks Discussion Honza Cernocky – next steps End 16.00, no buffet, drinks, entertainment  BUT Speech@FIT Honza Cernocky 05/201515/56

16. Program Honza Cernocky – intro, “why?” Ondrej Glembek – Technical description Petr Schwarz – Phonexia remarks Discussion Honza Cernocky – next steps End 16.00, no buffet, drinks, entertainment  BUT Speech@FIT Honza Cernocky 05/201516/56

17. Next steps If interested, sign-up to the google-group: http://groups.google.com/d/forum/voice-biometry- standard (no more personal emails).http://groups.google.com/d/forum/voice-biometry- standard take the code and test it on your data Report anything that you'd like to improve. Please bug-fixes, not complete changes … To the g-group or personally to Ondra glembek@fit.vutbr.cz glembek@fit.vutbr.cz Tell us if we can add your lab/company as supporter on the web-page. Please attach a logo in reasonable resolution and a web- link. You might need to consult your management. Vendors: implement it to your systems BUT Speech@FIT Honza Cernocky 05/201517/56

18. Ext steps II. The real normalization (ISO/IEC, NIST, W3C …) Yes, but only if it has wide industrial and academic support. Will need help … BUT Speech@FIT Honza Cernocky 05/201518/56

19. BUT Speech@FIT Honza Cernocky 05/201519/56 Thank you for your attention ! http://voicebiometry.org/