A SpeakUnique researcher is talking with a participant, who is wearing with headphones with a microphone piece. The researcher is describing the process for voice banking

SpeakUnique voicebank research

SpeakUnique voicebank research

Creating personalised voices for use in communications aids, simply and efficiently.

Sadly, MND can sometimes leave people unable to speak. When speech is impaired people often use a communication aid, such as an iPad or eye gaze machine, to speak for them. These devices come preinstalled with synthetic voices that are very generic, 'computerised' and unnatural sounding. They can leave people to feel that they have lost not just the ability to speak, but also a part of their identity. 

SpeakUnique, our Voicebank Research Project, was created to help people without the ability to speak to sound like themselves again, by providing ‘their own’ voice for use in a communication aid.

The research project is now complete and a spin-out company was launched in June 2020 and the products developed are available for use worldwide. Visit www.speakunique.co.uk to find out more about the company.

The technology

In collaboration with the Centre for Speech Technology Research and the Anne Rowling Regenerative Neurology Clinic at the University of Edinburgh, we developed technology that can create a synthetic version of an individual’s voice. These synthetic voices can then be integrated into the user’s communication aid, to allow them to communicate with a voice that is identifiably their own.

The technology we developed is unique in its ability to:

  1. Create a synthetic version of an individual’s own voice with a very short recording the user makes of their voice in their own home (around 40 minutes, compared to upwards of 6 hours from traditional speech synthesis technology);
  2. Repair speech where the user already has unclear speech, e.g. slurring or slowness, due to disease; this process can be further enhanced by mixing in the voice of a nominated voice donor (e.g. a sibling or childhood friend) who sounds similar.
  3. Design bespoke synthetic voices for those with no intelligible speech, based on desired characteristics such as age, gender and regional accent.

Our technology relies on accent templates that we created by blending together voice recordings of people with similar characteristics such as age and accent. During the project we recorded 1,644 donor voices from all over Scotland and other parts of the world. 

Thank you to everyone who donated their voice

Our Origins - SpeakUnique from SpeakUnique on Vimeo.