AI-Made Speech Teaches Machines to Hear Estonia and Slovenia
A new study shows that synthetic speech generated by large language models significantly improves speech recognition for low-resource Estonian and ...
A new study shows that synthetic speech generated by large language models significantly improves speech recognition for low-resource Estonian and ...
A new systematic review and meta-analysis finds that machine learning models can predict speech-in-noise performance with moderate to high accuracy, ...
Oregon State University researchers have built an AI-enabled smart compost bin that automatically weighs, images, and labels household food waste ...
New research in Nature Communications evaluates how well automatic speech recognition and large language models serve people with language impairments, ...
A new open dataset of 9300 Mandarin-English code-switched speech recordings, enhanced with AI-generated noise, speed and timbre changes, exposes how ...
An open-source pipeline called TAPA automates transcription, speaker diarization, forced alignment, and acoustic-phonetic measurement of naturalistic speech, validated against expert ...
© 2025 Scienmag - Science Magazine
© 2025 Scienmag - Science Magazine