Intron Unveils Sahara v2: Nigeria’s Voice AI Breakthrough Supporting 57 Languages and 500+ African Accents
A Nigerian artificial intelligence startup named Intron has officially announced an upgraded powerful speech recognition platform called Sahara v2, specifically designed for African languages and speech patterns. With this release, there will be features such as leap in voice AI technology, expanding support to 57 languages, including 23 African languages and more than 500 distinct accents across the continent.
In order to better capture real-world communication, Sahara v2 will be trained using over 14 million audio recordings obtained from over 40,000 speakers throughout Africa and the diaspora. Intron seeks to close the gap that occurs with main training in Western speech datasets, voice technologies created by foreign tech companies often mistake African names, phrases, and tone expressions.
The new development will attract newly added African languages such as Hausa, Yoruba, Igbo, Swahili, isiZulu, Kinyarwanda, Twi, Luganda, Wolof, Shona, Amharic, and African French. The development aims at introducing the world’s first Swahili-English automatic speech recognition (ASR) model designed to handle code-switching, a pattern which provides switching languages during discussion as one of the platform’s notable features.
The platform will be optimized in areas like healthcare, legal, financial services, and telecom, with improved recognition of African names, locations, and terminology.
About Intron
A Lagos-based AI business called Intron creates speech recognition software and voice technologies customized for African accents and languages. Tobi Olatunji and Olakunle Asekun launched the business in 2020 with a goal of constructing AI systems that are better able to fully understand the different languages of Africa.
Intron focuses on developing text-to-speech and speech-to-text infrastructure for usage in industries such as legal services, telecommunications, healthcare, and finance. Applications including voice banking, medical dictation, call center automation, and legal transcription are enabled by its technology, enabling businesses for precisely capture spoken conversations.