Markdown supported · ⌘↵ to post
Really interesting research direction! The point about AI voices breaking at Urdu-English switching moments highlights how much communication depends on cultural context, natural rhythm, and emotion not just pronunciation. I also like that the proposed study connects voice quality with actual purchase intent, making it relevant beyond technical performance. At Oglas AI, we believe practical AI works best when it reflects how people communicate and behave in the real world. Looking forward to seeing how this research develops!
Kartik N V J K
The 51.9 Hz versus 29.8 Hz pitch variability number is a good anchor for why a monolingual TTS voice goes flat right at "order now". I've heard the same stumble on Hindi-English lines, where the stress lands on the English word like it starts a new sentence. Are you planning to release the 50-script annotated corpus alongside the ABX results?