India Today on MSN

Meet world's first emotional speech-to-speech AI, Luna created by IIT-BHU alumnus

IIT-BHU alumnus Sparsh Agrawal has developed Luna, the world's first speech-to-speech foundational AI capable of singing, whispering, and responding with emotion. Built without big-tech backing, Luna ...
Newspoint on MSN

Jaipur's IIT student builds world's first speech-to-speech AI Model that can sing, whisper, feel

Jaipur-based 25-year-old founder Sparsh Agrawal has unveiled one of the first speech-to-speech foundational AI models that ...
marktechpost

Google Introduces Speech-to-Retrieval (S2R) Approach that Maps a Spoken Query Directly to an Embedding and Retrieves Information without First Converting Speech to Text

In the traditional cascade modeling approach, automatic speech recognition (ASR) first produces a single text string, which is then passed to retrieval. Small transcription errors can change query ...
IEEE

Bridging Modality Gap with Large Speech and Language Models for End-to-End Speech-to-Text Translation

Abstract: End-to-end speech-to-text translation (E2E ST) has increasingly aroused interest and attention recently, attempting to address the problem of data scarcity and modeling burden. Several ...
IEEE

StreamMel: Real-Time Zero-Shot Text-to-Speech Via Interleaved Continuous Autoregressive Modeling

Abstract: Recent advances in zero-shot text-to-speech (TTS) synthesis have achieved high-quality speech generationfor unseen speakers, but most systems remain unsuitable for real-time applications ...