Hugging Face's logo
Activity Feed

AI & ML interests

None defined yet.

Recent Activity
YannisTevissen  authored a paper 13 days ago
Beyond Anonymous Captions: Grounding Character Identity in Video Captioning and Question Answering
YannisTevissen  authored a paper about 1 month ago
Why Is Video Still So Expensive? A Survey of Inference-Efficiency Mechanisms in Video and Audiovisual LLMs
JunhaoZhuang  authored a paper about 2 months ago
EchoWM: Open and Enterable Omnimodal World Models
View all activity
Nymbo 
posted an update 5 days ago
view post
Post
3,622
Introducing Live Subtitles, a Windows tray app that captions anything your PC is playing and isn't butt ugly. live and fully local. It runs NVIDIA's Parakeet-unified 0.6B on your GPU, puts words on screen about 0.36 s after they're spoken, and uses less than one CPU core while it does it.

I built a benchmark for it out of real TV, podcasts, earnings calls and a short film. Parakeet-unified scores 12.1% WER there, close to Nemotron streaming's 11.4%, but with much better punctuation and lower worst-case delay. Once a caption is on screen it never gets rewritten or jumps lines, and the overlay disappears when nobody's talking.

No NVIDIA GPU? It falls back to Nemotron on the CPU, which also covers 30 other languages. It can also write SRT/ASS/VTT files for videos on disk.

Windows only. The GPU engine needs an NVIDIA card with ~3.2 GB of free VRAM. MIT.

Edit: the first version of this post said 13.8% vs 17.0%. The replies caught a scoring bug (numbers weren't normalized), so the benchmark now uses Whisper's English normalizer and the numbers above are re-scored.

Get it here:
https://github.com/Nymbo/Live-Subtitles

Models:
nvidia/parakeet-unified-en-0.6b
nvidia/nemotron-speech-streaming-en-0.6b
  • 7 replies
·
aotrih 
authored a paper 14 days ago
aotrih 
authored a paper about 1 month ago
mahwizzzz 
posted an update about 1 month ago
view post
Post
235
I implemented the attention-free bidirectional encoder architecture Avey-B for Urdu a compact 24.87M-parameter language encoder built for efficient Urdu NLP research.

Original Avey-B paper: Avey-B (2602.15814)
Urdu model: mahwizzzz/avey-b-ur
Michielo 
in social-post-explorers/README 2 months ago

question on how to join and etc

9
#49 opened 6 months ago by
simonko912
SarosVale 
in social-post-explorers/README 2 months ago

Request to join group

7
#55 opened 2 months ago by
NILKNARFGonzo
mahwizzzz 
posted an update 2 months ago
andreasmartin 
submitted a paper to Daily Papers 2 months ago
Nymbo 
posted an update 2 months ago
view post
Post
2,539
Anthropic gave me six months of Claude Max 20x through the Claude for Open Source program, granted based on my Hugging Face work. Thank you
Anthropic for supporting open source.

So far I've been pointing it at Markdown Minimap, an Obsidian plugin that adds a scrollable IDE-style minimap to your notes. This week I've been clearing a backlog of user-reported issues on it, with Claude often handling them end to end.

https://github.com/Nymbo/Markdown-Minimap — issues and PRs welcome.
gagan3012 
authored a paper 3 months ago
gagan3012 
submitted a paper to Daily Papers 3 months ago
ucheog 
in social-post-explorers/README 3 months ago

Question

2
#52 opened 3 months ago by
simonko912

Membership request — stevenpr (open-source local-LLM app & model author)

🔥 2
1
#51 opened 3 months ago by
stevenpr
Nymbo 
posted an update 3 months ago
view post
Post
6,191
Introducing Inflect-v2, two exceptionally small, open-weight English TTS models at just 3.9M and 9.3M parameters. Both generate speech multiple times faster than real-time on CPU. Despite their size, Inflect-v2 delivers quality that is competitive with much larger lightweight TTS systems, including KittenTTS, Piper, and Supertonic-3.

CPU, CUDA, PyTorch, and ONNX are supported. Apache 2.0.

See it for yourselves:
owensong/Inflect-Micro-v2
owensong/Inflect-Nano-v2

Try the Demos:
Nymbo/Inflect-TTS (unlimited CPU usage)
owensong/Inflect-v2 (ultra-fast ZeroGPU usage)
  • 6 replies
·
codelion 
authored 4 papers 3 months ago
codelion 
posted an update 4 months ago
view post
Post
804
SPROG-9M — a 9.37M parameter model trained from scratch to solve GSM8K-style math without using an LLM at inference.

The model, codelion/sprog-9m, predicts symbolic programs over number slots, then a deterministic executor does the arithmetic. With a simple verifier, it reaches ~11.8% on GSM8K test.

We also released the dataset: codelion/gsm8k-synth, 117K validated synthetic GSM8K-style problems.

Tiny model, no pretraining, no LLM at inference, runs on a laptop.
gagan3012 
authored a paper 5 months ago