Tootfinder

1/ Feels a bit poignant to be putting finishing touches on the next vesion of my book (PLAI 3.2.5) knowing it'll be the last of the v3, and if my planned experiment goes well, it may be the last longform book I will ever write. I will truly miss longform writing: ↵

Enabling automatic transcription of child-centered audio recordings from real-world environments
Daniil Kocharov, Okko R\"as\"anen
https://arxiv.org/abs/2506.11747

Enabling automatic transcription of child-centered audio recordings from real-world environments
Longform audio recordings obtained with microphones worn by children-also known as child-centered daylong recordings-have become a standard method for studying children's language experiences and their impact on subsequent language development. Transcripts of longform speech audio would enable rich analyses at various linguistic levels, yet the massive scale of typical longform corpora prohibits comprehensive manual annotation. At the same time, automatic speech recognition (ASR)-based transcri…

Consider this - "only 10.6 per cent of #India’s population speaks #English"
Yet so many major #Indian media outlets only publish in English, especially the ones doing investigative of longform journali…