Meta recently introduced its multimodal AI translation model, SeamlessM4T, which supports almost 100 languages for text and 36 for speech. Now, with an updated “v2” architecture, the company is expanding on this tool to enhance conversational translations, making them more spontaneous and expressive. This is a crucial element for achieving authentic conversations across different languages.
One of the new features is called “SeamlessExpressive,” which allows your expressions to be conveyed in your translated speech. This includes your pitch, volume, emotional tone (such as excitement, sadness, or whispers), speech rate, and pauses. This breakthrough has the potential to revolutionize our daily lives and content production, as translated speeches have often sounded robotic in the past. The supported languages for this feature include English, Spanish, German, French, Italian, and Chinese. However, please note that the demo page currently does not include Italian and Chinese.
The second…
2023-12-03 22:50:14
Source from www.engadget.com