Synthetic voice is now able to read Kazakh text – a new development of NU scientists

A computer voice will now be able to read Kazakh text. This was made possible thanks to a new project of scientists from the Institute of Smart Systems and Artificial Intelligence (ISSAI) of Nazarbayev University. They have developed a Kazakh language speech synthesis project, or in other words the Kazakh text-to-speech conversion. Text-to-speech conversion is the artificial production of human speech which allows a computer to read any text in a voice that is as close to natural as possible.

“In Kazakhstan this technology can be used for people with special needs, virtual assistants, and the public services where text-to-speech conversion is necessary. This technology is an essential component in many applications such as interactive smart assistant systems, navigation systems, announcement systems where computer voice-over of the Kazakh text is required. The technology can also be useful for visually impaired people, as it will allow to establish human-robot interaction by converting Kazakh text into audio”, – says ISSAI postdoctoral student Yerbolat Khasanov, – “One of the main goals of our project was to develop the use of the Kazakh language in the digital sphere. Before, we did not meet such ready-made Kazakh speech synthesis projects”.

Scientists have created two versions of the computer voice – male and female, which can voice Kazakh text with the correct intonation and tempo.

Worldwide, this technology is widely used in such industries as healthcare, e-learning, and voice assistants.

Scientists share the collected database, pretrained models and project code in an open-source format. With this, any organization or company will be able to create their own projects related to Kazakh text-to-speech conversion.

Demo version of the kazakh text-to-speech technology can be tested via the link: https://issai.nu.edu.kz/tts

Videos with male and female voices are available via the link: https://issai.nu.edu.kz/tts