Forensic Data Science Laboratory
Home	E3 Forensic Speech Science System (E3FS3)

Description

• The E3 Forensic Speech Science System (E³FS³) is designed for forensic-voice-comparison research and casework practice. It is designed:
  
  
  • to assess strength of evidence as a numeric likelihood ratio, in accordance with modern thinking on interpretation of forensic evidence;
    
    
  • to reduce the potential for cognitive bias, via the use of relevant data, quantitative measurements, and statistical models, with subjective judgement kept as far away as possible from the strength-of-evidence conclusion; and
    
    
  • to facilitate and survive rigorous examination under legal admissibility standards such as those of US Federal Rule of Evidence 702 and the Supreme Court rulings in Daubert, Joiner, and Kumho Tire, and of England & Wales Criminal Practice Directions 19A.
    
    
• Using “system” in a broad sense, the system will include:
  
  
  • software tools;
    
    
  • databases (and software tools and protocols for data collection);
    
    
  • standards, guidelines, and standard operating procedures (SOPs);
    
    
  • a library of validation reports; and
    
    
  • training for practitioners.
    
    
• The software tools implement state-of-the-art automatic-speaker-recognition algorithms, and are accompanied by detailed documentation explaining which algorithms were implemented and why they were chosen. For maximum transparency, the software code is extensively commented and is open for inspection by the court (open code). The software tools are designed to make it easy for researchers and end-users to conduct validation studies.
  
  
  • 2019-08 – 2020-03. End-user-needs assessments were conducted.
    
    
  • 2020-01 – 2021-07. An alpha version of core software tools was developed and validated.
    
    
  • 2021-08. A partner organization completed the first casework report using the alpha version of the core software tools.
    
    
  • 2021-09 – 2022-12. A beta version of the software tools was developed.
    
    
  • Pending: Field testing of beta version by partner organizations.
    
    
• The databases consist of audio recordings of speakers, and represent relevant populations and reflect conditions commonly encountered by end-users in casework.
  
  
  • 2019-08 – 2020-03. End-user-needs assessments were conducted.
    
    
  • 2020-09 – present. Software tools to facilitate data collection are under development. The software tools include a smartphone app that helps schedule remote recording tasks and on-site visits, and that automatically uploads remotely made recordings to a server.
    
    
  • Due to the COVID-19 pandemic, no new data collection has been conducted.
    
    
• The standards and guidelines provide requirements and recommendations on how practitioners should conduct evaluations and validations. End-users who wish to use the system for casework are required to agree to follow designated standards and guidelines. Example SOPs provide detailed instructions on how to use the software tools and databases in a manner that is in accordance with the designated standards and guidelines.
  
  
  • 2019-09 – 2020-08. We developed the guideline: Consensus on validation of forensic voice comparison.
    
    
  • 2021-07. A checklist for conformity with the Consensus was developed.
    
    
  • 2018-08 – present. We are contributing to the development of ISO 21043 Forensic Science.
    
    
• The library of validation reports demonstrates how well the software tools and procedures (“system” in a narrow sense) perform under a range of casework conditions.
  
  
  • 2021-07 – 2021-12. We conducted multiple validations of core software tools under several sets of conditions and prepared a validation report.
    
    
  • Pending: Additional validation studies.
    
    
• The training covers forensic inference and statistics, including understanding of the likelihood-ratio framework, of calibration and validation, and of cognitive bias. It also covers forensic voice comparison in particular, including understanding of statistical models used in human-supervised-automatic approaches to forensic voice comparison, and practical use of the system’s software tools.
  
  
  • We plan to offer an online interactive continuing-professional-development (CPD) for-credit programme. The first course will be on forensic inference and statistics, with expected first delivery beginning 2023-09. The second course will be on forensic voice comparison with expected first delivery beginning 2024-09.
    
    

Selected Publications

Morrison G.S., Zhang C. (2023). Forensic voice comparison: Overview. In Houck M., Wilson L., Lewis S., Eldridge H., Lothridge K., Reedy P. (Eds.), Encyclopedia of Forensic Sciences (3rd Ed.), vol. 2, pp. 737–750. Elsevier.
https://doi.org/10.1016/B978-0-12-823677-2.00130-6
Preprint: http://forensic-voice-comparison.net/encyclopedia/



Morrison G.S., Weber P., Enzinger E., Labrador B., Lozano-Díez A., Ramos D., González-Rodríguez J., (2023). Forensic voice comparison: Human-supervised-automatic approach. In Houck M., Wilson L., Lewis S., Eldridge H., Lothridge K., Reedy P. (Eds.), Encyclopedia of Forensic Sciences (3rd Ed.), vol. 2, pp. 720–736. Elsevier.
https://doi.org/10.1016/B978-0-12-823677-2.00182-3
Preprint: http://forensic-voice-comparison.net/encyclopedia/



Weber P., Enzinger E., Labrador B., Lozano-Díez A., Ramos D., González-Rodríguez J., Morrison G.S. (2022). Validation of the alpha version of the E³ Forensic Speech Science System (E³FS³) core software tools. Forensic Science International: Syenergy, 4, article 100223.
https://doi.org/10.1016/j.fsisyn.2022.100223

Morrison G.S., Enzinger E., Hughes V., Jessen M., Meuwly D., Neumann C., Planting S., Thompson W.C., van der Vloed D., Ypma R.J.F., Zhang C., Anonymous A., Anonymous B. (2021). Consensus on validation of forensic voice comparison. Science & Justice, 61, 229–309.
https://doi.org/10.1016/j.scijus.2021.02.002



Morrison G.S., Enzinger E., Ramos D., González-Rodríguez J., Lozano-Díez A. (2020). Statistical models in forensic voice comparison. In Banks D.L., Kafadar K., Kaye D.H., Tackett M. (Eds.), Handbook of Forensic Statistics (Ch. 20, pp. 451–497). Boca Raton, FL: CRC.
https://doi.org/10.1201/9780367527709
Preprint: http://handbook-of-forensic-statistics.forensic-voice-comparison.net/



Morrison G.S., Enzinger E. (2019). Introduction to forensic voice comparison. In Katz W.F., Assmann P.F. (Eds.) The Routledge Handbook of Phonetics (ch. 21, pp. 599–634). Abingdon, UK: Taylor & Francis.
https://doi.org/10.4324/9780429056253-22
Preprint: https://research.aston.ac.uk/en/publications/introduction-to-forensic-voice-comparison



Morrison G.S., Kelly F. (2019). A statistical procedure to adjust for time-interval mismatch in forensic voice comparison. Speech Communication, 112, 15–21.
https://doi.org/10.1016/j.specom.2019.07.001



Morrison G.S., Enzinger E., Zhang C. (2018). Forensic speech science. In Freckelton I., Selby H., (Eds.), Expert Evidence (Ch. 99). Sydney, Australia: Thomson Reuters.
Preprint: http://expert-evidence.forensic-voice-comparison.net/



Morrison G.S. (2018). Admissibility of forensic voice comparison testimony in England and Wales. Criminal Law Review, 2018(1), 20–33.
Preprint: http://geoff-morrison.net/#Admissibility_EW_2018



Morrison G.S., Thompson W.C. (2017). Assessing the admissibility of a new generation of forensic voice comparison testimony. Columbia Science and Technology Law Review, 18, 326–434.
https://doi.org/10.7916/stlr.v18i2.4022
Preprint: https://ssrn.com/abstract=2883767
Preprint: https://www.newton.ac.uk/files/preprints/ni16053.pdf

Documentation

– Technical report on design and implementation of software tools

Sorry, nothing publicly released yet. Please check back at a later date.

– Data-collection protocols

Sorry, nothing publicly released yet. Please check back at a later date.

Software

– E³FS³ software components

Sorry, nothing publicly released yet. Please check back at a later date.

Databases

– Descriptions and links to databases:

http://databases.forensic-voice-comparison.net/



Due to the COVID-19 pandemic, no new data collection has been conducted.



We plan to distribute databases through the Linguistic Data Consortium.

Links

– Forensic voice comparison portal

– 2016–2019 Virtual Special Issue of Speech Communication:

Multi-laboratory evaluation of forensic voice comparison systems under conditions reflecting those of a real forensic case (forensic_eval_01)

http://forensic-voice-comparison.net/E3FS3/

This webpage is maintained by Geoffrey Stewart Morrison and was last updated 2023-03-16