Tootfinder

Simul-Whisper: Attention-Guided Streaming Whisper with Truncation Detection
Haoyu Wang, Guoqiang Hu, Guodong Lin, Wei-Qiang Zhang, Jian Li
https://arxiv.org/abs/2406.10052

Simul-Whisper: Attention-Guided Streaming Whisper with Truncation Detection
As a robust and large-scale multilingual speech recognition model, Whisper has demonstrated impressive results in many low-resource and out-of-distribution scenarios. However, its encoder-decoder structure hinders its application to streaming speech recognition. In this paper, we introduce Simul-Whisper, which uses the time alignment embedded in Whisper's cross-attention to guide auto-regressive decoding and achieve chunk-based streaming ASR without any fine-tuning of the pre-trained model. Fur…

Oooh, ooh, ooh… look what just arrived! I’ve been waiting almost a year for it and boy was it worth the wait. My new Linux laptop is a… tablet? Yep. The StarLabs StarLite.
Oh and the screen on this thing! 😍
https://ie.starlabs.systems/pages/starlite

StarLite 12.5-inch
Its fanless design ensures your StarLite will never make more than a whisper - unless you want it to. The Mk V supports coreboot open source firmware which you can effortlessly configure to your preferences via our coreboot configurator.

gently whispering
Revert "Revert "Revert "[Common] Simplify. NFC"""
in a crowded room and seeing every LLVM developer flinch and shudder as if suddenly overcome with extreme pain

So which one of you... *whispers* wears the cat ears in the relationship?

Whisper-Flamingo: Integrating Visual Features into Whisper for Audio-Visual Speech Recognition and Translation
Andrew Rouditchenko, Yuan Gong, Samuel Thomas, Leonid Karlinsky, Hilde Kuehne, Rogerio Feris, James Glass
https://arxiv.org/abs/2406.10082

Whisper-Flamingo: Integrating Visual Features into Whisper for Audio-Visual Speech Recognition and Translation
Audio-Visual Speech Recognition (AVSR) uses lip-based video to improve performance in noise. Since videos are harder to obtain than audio, the video training data of AVSR models is usually limited to a few thousand hours. In contrast, speech models such as Whisper are trained with hundreds of thousands of hours of data, and thus learn a better speech-to-text decoder. The huge training data difference motivates us to adapt Whisper to handle video inputs. Inspired by Flamingo which injects visual…

I'm reading The Song of Achilles, which is either a gay romance novel or else a scholarly retelling of The Iliad or both. The prose style is interesting, lyrical and just a bit foreign. It feels like reading translations of Homer only modern and much easier to follow. I'm not normally one for wrought prose but this works to transport me.
Their whispers choked me, turned the food in my mouth to ash. I pushed away my plate and sought out corners and spare halls where I might sit undisturbed, except for the occasional passing servant. My narrow world narrowed further: to the cracks in the floor, the carved whorls in the stone walls. They rasped softly as I traced them with my fingertip.
“I heard you were here.” A clear voice, like ice-melted streams.
My head jerked up. I was in a storeroom, my knees against my chest, wedged between jars of thick-pressed olive oil. I had been dreaming myself a fish, silvered by sun as it leapt from the sea. The waves dissolved, became amphorae and grain sacks again.