Forensic voice comparison – VoxCrim

The project defines the concept of «box-rule«, a restricted subspace where a reliable forensic voice comparison is possible.
For that, it is mandatory to describe the acoustic space with speaker specific phonetic information.
The project will also propose a method able to objectively evaluate the tying between a real case and a given boxrule, using a representation learning (deep learning) approach.
The final aim is to propose an objective standard inspired from ISO 17025.

A complete toolbox issued from FABIOLE project and adapted to voxCrim was realizeed
A more realistic database (PTSVox) was designed and the first part recorded.
A first study of phonetic speaker specific information was completed

To build FABIOLE II, a databse allowing large scale intraspeaker variability studies.
Extention of PTSVox
To propose a first boxrule with the associated standard checking protocols.

Nautsch, A., Jiménez, A., Treiber, A., Kolberg, J., Jasserand, C., Kindt, E., ..., Bonastre J.-F. & Abdelraheem, M. A. (2019). Preserving Privacy in Speaker and Speech Characterisation. Computer Speech & Language.
Audibert, N., Fougeron, C. (2019). Are people as old as they sound? Acoustic, regional and generational effects. Proceedings of the 3rd Phonetics and Phonology in Europe conference, Lecce, Italy, pp. 49-50.
Ferragne, E., Gendrot, C., & Pellegrini, T. (2019). Towards Phonetic Interpretability in Deep Learning Applied to Voice Comparison. ICPhS, Melbourne.
Gendrot, C., Ferragne, E. & Pellegrini, T. (2019). Deep learning and voice comparison: phonetically-motivated vs. automatically-learned features
Georgeton, L. (2019) Project Voxcrim : Forensic Voice Comparison, IAFPA, Istanbul.
Lapidot, I., Delgado, H., Todisco, M., Evans, N. W., & Bonastre, J. F. (2018). Speech Database and Protocol Validation Using Waveform Entropy. In Interspeech (pp. 2773-2777).
Ajili, M., Bonastre, J. F., & Rossato, S. (2018). Voice Comparison and Rhythm: Behavioral Differences between Target and Non-target Comparisons. Proc. Interspeech 2018, 1061-1065.
Audibert, N., Fougeron, C., Brunot, M., Callé, S., Davallet, I., Lechevalier, N. (2019). Liens entre âge estimé, sévérité de la dysarthrie et intelligibilité des locuteurs. Actes des 8èmes Journées de Phonétique Clinique, Mons, Belgique, pp. 19-22.
Ajili, M., Bonastre, J. F., Waad, B. K., Rossato, S., & Kahn, J. (2018). Comparaison des voix dans le cadre judiciaire: influence du contenu phonétique. Proc. XXXIIe Journées d’Études sur la Parole, 28-36.

VoxCrim deals with speaker authentication and more precisely forensic voice comparison. VoxCrim targets national security and legal/justice application contexts. The project's aims correspond to "ANR Défi 9". VoxCrim proposes a validated scientific objective framework available for all kinds of forensic voice comparison methods (automatic or phonetic). The goal is to develop certified standards to delimit the specific areas where voice comparison methods are applicable. VoxCrim gathers computer scientists (LIA) with phoneticians (LPL, LPP), experts in standardization (LNE) and members of the forensic department of Police Nationale (SDPTS). Members of the forensics department of Gendarmerie Nationale will take part in the project as a center of expertise. VoxCrim's aims are organised along two time scales: (1) analyses of well-controlled recordings in the near future, and (2) an enlarged area of application later on. (1) VoxCrim proposes to develop methods and databases based on the results of ANR Fabiole in order to extend the innovative concept of "box rule" (i.e. a set of conditions where voice comparison is perfectly applicable and where certification is possible). (2) Then, the project aims to add dimensions to the standards of voice comparison, such as the influence of the socio-cultural and linguistic environments. It is necessary to add these characteristics of the voice because they are closely linked to the real-life context police services are faced with. For this purpose, a database (PTVVox, with two recordings conditions i.e micro and GSM conditions) recorded by the SDPTS has been created and will be exploited. Two types of complementary analyses will be conducted: a) an acoustic analysis where relevant cues will be extracted for voice characterization and b) perceptual experiments where the ability of listeners to discriminate voices will be tested. Acoustic analyses will contribute to delimit inter and intra speaker variability in order to test the robustness of these cues. Perceptual experiments will verify if the acoustic cues previously highlighted are actually used by listeners to identify speakers. These experiments will be conducted particularly in GSM condition (corresponding to the most common case in forensic cases). This project will allow the SDPTS labs to extend their ISO 17025 validation approaches to voice comparison. In order to disseminate the knowledge and the questions about voice comparison, seminars will be organized with the speech community and members of the judicial system. The project will also help to gather – and provide training for – a pool of young researchers specializing in voice comparison and the forensic applications thereof in order to compensate for a well-known lack of specialists in France.