Translation. Region: Russian Federation –
Source: People's Republic of China in Russian – People's Republic of China in Russian –
An important disclaimer is at the bottom of this article.
Source: People's Republic of China – State Council News
Astana, January 18 (Xinhua) — Kazakhstan has developed its first large-scale automatic speech recognition (ASR) model for Turkic languages, making it the largest such development in Central Asia. This was reported on Sunday by the Kazinform news agency, citing the country's Ministry of Artificial Intelligence and Digital Development.
The new AI model is designed with the linguistic characteristics of the Turkic language group in mind and is not an adaptation of English-language solutions, which distinguishes it from technologies previously used in the region. The project was implemented with the support of the Astana Hub and the relevant ministry, using Microsoft's GPU infrastructure.
The model is reportedly trained in Kazakh, Turkish, Uzbek, Kyrgyz, Azerbaijani, and Tatar languages, and is also capable of correctly recognizing mixed Turkic-Russian speech, which is widely used in everyday communications, business environments, and in the provision of public services.
The developers note that the new ASR model is robust to noise, accents, and language switching within a single utterance, and demonstrates high accuracy across a variety of applications, from financial services and telecommunications to logistics and e-government.
According to the agency, the implementation of this technology will reduce costs in customer support systems and ensure 24/7 availability of voice services in national languages. –0–
Please note: This information is raw content obtained directly from the source. It represents an accurate account of the source's assertions and does not necessarily reflect the position of MIL-OSI or its clients.