Empowering the Vocally Impaired With AI-Generated, Expressive Voices

Translated is proud to be the tech partner of a project that enables those with voice impairments to communicate using expressive voices. The initiative has received the endorsement of the World Health Organization Director General Tedros Adhanom Ghebreyesus and the Italian Prime Minister Giorgia Meloni.

Rome – February 6, 2023

Translated is excited to announce a partnership with Nemo Lab, a technology center for neuromuscular disease research, and Campus Bio-Medico University of Rome, a university focused on science and technology for human wellness, on a project named Voice For Purpose. The goal of the project is to enable individuals who use augmentative and alternative communication (AAC) devices to communicate with expressive voices.

The initiative was unveiled today at the Italian headquarters of the European Parliament, with messages from the World Health Organization Director General Tedros Adhanom Ghebreyesus and the Italian Prime Minister Giorgia Meloni, and the participation of the Director General for Health and Food Safety of the European Commission, Sandra Gallina, and the Italian Minister of Health, Orazio Schillaci.

Translated is honored to serve as the technology partner on the project, which offers a crucial lifeline to those with speech impairments. The company has developed a web portal, currently available in five languages, where users can record and donate their voices or request a voice model based on features like age, gender, and language.

This initiative was made possible thanks to recent advances in AI voice generation, which Translated has pioneered to provide high-quality audiovisual localization services on a large scale. The voice models are trained by Translated's cutting-edge neural synthesis engine to produce speech that has natural prosody, rhythm, and tone. This technology has already been used in dubbing projects and can generate expressive AI voices using less than an hour of recordings. This is especially crucial for those whose voices are already compromised and who find it difficult to speak for long periods. The voice model can be used with AAC devices, smartphones, and computers.

It’s amazing to see how the AI voices we develop help people communicate with each other. Making this project work requires the capacity to scale AI voices and serve the needs of people in different conditions. We learn how to improve by collaborating with patients every day. We are proud to offer our knowledge and computing power to this project, which contributes to the quality of people’s lives.
Fabio Minazzi – Director of Audiovisual

Voice For Purpose relies on a customized version of Translated's human-centric technology, which was designed to support professional translators. A pilot program was run before releasing the web platform, which included collaborators AISLA, the Italian association for ALS, and AAC device manufacturers, and which saw the participation of 12 users and over 250 donors.

Translated is calling on everyone to donate their voice to the Voice For Purpose library. The process of recording a sample for the voice library only takes a few minutes. By having a voice that is more like their own, those with speech impairments will be able to communicate and participate in social interactions more effectively and naturally when speaking to their caregivers, or simply say “I love you” to their loved ones.

Support the project

Record a sample of your voice and make it available in the Voiceforpurpose.com voice library.

Donate your voice