Common Voice added the Tajik language to develop conversational AI

The Tajik language has appeared on the Common Voice platform, marking an important step toward the development of conversational artificial intelligence. Now, work begins on creating the largest open dataset for Tajik speech.

This success was made possible by translating the platform's interface into Tajik. Initially, the project started as a personal initiative, but later efforts were combined with community participants. A breakthrough was achieved after a presentation at DevFest GDG Khujand, where the idea found support.

The next stage is collecting Tajik language sentences to create training materials. Volunteers will be able to record their voices to train speech recognition algorithms.

Importantly, the project is entirely volunteer-driven and focused on openness: the collected data will be available to everyone, including researchers, educational institutions, and even commercial projects. This approach aims to address data privacy issues in scientific research and accelerate the development of algorithms for Tajik speech recognition and voice synthesis.

Source: rational_optimist