3 papers
cs.CV2026
STEPS: Scene Text Editing with Preserved Style Using Diffusion and Contrastive Style Encoding
Nicolas Thiebaut, Nameer Hirschkind, Xiao Yu +1
We introduce Scene Text Editing with Preserved Style (STEPS), a novel diffusion model architecture for quality text replacement in images. Scene Text Editing (STE), also known as V…
cs.LG2024
Diffusion Synthesizer for Efficient Multilingual Speech to Speech Translation
Nameer Hirschkind, Xiao Yu, Mahesh Kumar Nandwana +9
We introduce DiffuseST, a low-latency, direct speech-to-speech translation system capable of preserving the input speaker's voice zero-shot while translating from multiple source l…
cs.CL2023
Learning When to Speak: Latency and Quality Trade-offs for Simultaneous Speech-to-Speech Translation with Offline Models
Liam Dugan, Anshul Wadhawan, Kyle Spence +3
Recent work in speech-to-speech translation (S2ST) has focused primarily on offline settings, where the full input utterance is available before any output is given. This, however,…