Imagine an application that can modify the emotional tone of your voice to make you feel happier… This device actually already exists! It was designed by researchers from the CNRS (France’s National Scientific Research Council), Japanese, Swedish and French universities. Their study shows that our brain regulates mood according to the tone of our own voice. So how can we purposefully manipulate our voices to change our emotional state?

We are capable of controlling and regulating emotional expression by trying, for example, to appear unaffected by an event. The studies blurred the boundaries between cognitive and emotional processes. Thus, according to J-J. Aucouturier, the lead author of study, “we ought to be able to commit emotional errors, detect them, and correct them.” But to better understand how conscience people are of their own emotional expressions, vocal expression is undoubtedly the best candidate. Indeed, in the instant between speaking and hearing one's voice, the vocal signals exist for a brief moment outside of the body's sensory circuits. This means that it’s possible de capture the voice signal and to modify the emotional tone before sending it back to the speaker. In this case, is the speaker aware of the manipulation? To address the question, J-J. Aucouturier of IRCAM (France’s Institute of Musical/Acoustic Research and Coordination) and colleagues decided to create a digital audio platform for studying the mechanisms behind the production of voice emotion.

In this first study published in PNAS (Proceedings of the National Academy of Science), participants were asked to read a short text out loud (an excerpt from the short story, The Second Bakery Attack by Haruki Murakami, translated into French) while listening to the sound of their voice through headphones. Without their knowledge, their voices were slightly modified to sound happier, sadder, or fearful. How did the researchers emotionally modify the voices? We know that a happy tone can be recreated by increasing the voice pitch by a third (or fourth) and by emphasizing the first syllable of each word. Sadness on the other hand can be artificially produced by decreasing pitch and by softening the ends of words. The voices of the participants were modified using digital audio signal processing algorithms.

For this research, 112 participants (92 women, average age: 20.1) answered a series of questions in order to determine whether they had consciously detected that their voices had been manipulated. 90 other participants (all women; average age: 21) reported their emotional states both before and after the reading task. Their responses were grouped according to emotion (happy, optimistic and sad) and tension (upset, anxious and relaxed). The results showed that when participants listened to their voice, their emotional state changed in the direction of the modification (happy, sad, or fearful) and that they did not perceive (in the vast majority of cases) the change in their voice. According to Aucouturier, the study shows that the brain automatically processes the sound of one’s own voice, without our knowledge, and informs us about our emotional state.

Practically speaking, this means that it's possible to influence our emotional state by identifying voice characteristics and adjusting accordingly in order to trigger mood change in the brain. K. Watanabe, one of the study’s co-authors, believes that the platform could be used for therapeutic purposes. If you want to try out this “tool," go to the following site: http://cream.ircam.fr/?p=44; it’s free!