Exploring representation, identity and aesthetics of synthesized voices

What are the politics and aesthetics involved in synthesized voices? What happens during the listening experience when the voice we hear is synthesized? What are the notions of identity, personhood and will that we ascribe to the synthesized voices talking to us?

Synthetic voices are artificial voices generated by algorithms. Currently, the designed synthesized voices that we hear in our everyday life seem to have only one vocal identity. In multivocal we question the aesthetic design as well as the representational modes of these synthesized voices, and ask: since voices from machines are not limited to a single vocal identity, why do the currently available synthesized voices have only one gender, one age, and one accent?

multivocal aims to produce reflections on the implications resulting from a single synthesized voice identity, and create new listening experiences to improve and reimagine human and machine relationships.

> Recent blog posts

Talk and installation at Driving IT

The voice-recording box is now up at IDA on Kalvebodbrygge 31 in Copenhagen. We will here collect voice recordings for the next weeks. In additions to this, Frederik is doing a talk at 15:00 at the Driving IT conference.


Interview with IDA and Henrik Føhns

Stina and Frederik did an interview with Henrik Føhns at Techfestival. This was done for the IDA podcast called Techtopida. The interview can be found here, it starts at 19:10 and is in Danish.


Events this week

Voice box is up and we are doing two talks!


The multivocal voice of Roskilde


Website is live!