< Back to IRCAM Forum

Welcome to the DAVID user-group!

DAVID
#1

DAVID (Da Amazing Voice Inflection Device) is a free, real-time voice transformation tool able to “colour” any voice recording with an emotion that wasn’t intended by its speaker. DAVID was especially designed with the affective psychology and neuroscience community in mind, and aims to provide researchers with new ways to produce and control affective stimuli, both for offline listening and for real-time paradigms.

DAVID was originally developped by the CREAM lab in IRCAM/CNRS to support research on emotional vocal feedback (Aucouturier, J.J., Johansson, P., Hall, L., Segnini, R., Mercadié, L. & Watanabe, K. (2016) Covert Digital Manipulation of Vocal Emotion Alter Speakers’ Emotional State in a Congruent Direction. Proceedings of the National Academy of Sciences, vol. 113 no. 4, doi: 10.1073/pnas.1506552113). Follow-up experiments around the same paradigm are ongoing at our collaborators’ labs in Paris, Lund, London and Tokyo. Feel free to get in touch using this user-group if you are planning to use DAVID for your own work.

DAVID was released as v1.0 in October 2015, and has seen 2 major updates since then. We are continuously working to improve its ease-of-use, functionalities, and to fix potential technical issues. Please do report any questions or issue you may have using the software, and we’ll try to get in touch with you asap.

Finally, just a quick word about future updates that we are working on for the software, so you now what’s in store:

  • A number of projects have started to use DAVID as a way to manipulate emotional expression on a trial-per-trial basis. The project now includes a python wrapper, that can control the DAVID patch and change presets on and off to match whatever experimental presentation is needed. This work well with experimental scripst written in Python/Psychopy, for instance. We will soon post an announcement about the release of that code.
  • We are also working on adapting DAVID to ERP experiments, e.g. to investigate error monitoring in speech production. New upcoming versions of DAVID will include the possibility to record time stamps that mark the onset of voice manipulations, so brain potentials can be time-aligned with the events. Again, watch this place for an announcement about this feature as soon as this is ready.

Here it is. This was built with love, we use it everyday, and we’d love to hear about you.

JJ Aucouturier, Marco Liuni
Cream Lab, CNRS/IRCAM
http://cream.ircam.fr

#2

Dear JJ Aucouturier, Marco Liuni,
I was wondering if I input a wave file with someone’s voice would the software tell me the emotions. For example the person was afraid, sad, happy etc.
The application that I am working on is to detect the emotions of a person by asking the user to record a sentence and I would like to input that wav file to DAVID software and would like to get an output such as sad, happy etc. Is this possible? or any ideas that could suggest. Thanks in advance for your help.
Kamal

#3

Hi Jean-Julien

Hope this email finds you well,

I’m a PhD candidate from University of Warwick and my name is Junning. Thank you for all the hard works in developing DAVID and other voice-emotion related software. Currently, I have an ongoing project which is investigating how human vocal expression would affect their perceived trustworthiness, and I’m planning to use DAVID to manipulate our participants’ voice in more controlled directions.

I initially found out about this software from the IRCAM Forum, thus, I’m wondering whether there has been more updated DAVID version which offers more control on voice features or has more default emotional settings? If so, would it be possible for you to share the latest version of DAVID?

Thank you so much for your time and help,

All the best,
Junning