American tech company Meta, the parent company of Facebook, has recently launched an AI model for translating speech and texts called SeamlessM4T. This model can reportedly translate speech to speech, speech to text, text to speech, and text to text in nearly 100 languages.
According to the Meta, SeamlessM4T is the world’s “first all-in-one multimodal and multilingual AI translation model”. This is in keeping with the company’s effort to make AI tools with multilingual capabilities.
“SeamlessM4T builds on advancements we and others have made over the years in the quest to create a universal translator. Compared to approaches using separate models, SeamlessM4T’s single system approach reduces errors and delays, increasing the efficiency and quality of the translation process”, the company announced in a press release.
The company has also decided to publicly release the AI model under a research license as well as releasing the metadata SeamlessAlign, which has over 270,000 hours of mined speech and text alignments.
Describing the thought process of making the software public, the company stated, “We’re publicly releasing SeamlessM4T under a research license to allow researchers and developers to build on this work.”.