Heriot-Watt University

classroom

https://www.hw.ac.uk/

Activity Feed Request to join this org

AI & ML interests

Machine Learning, Dialogue Systems, Natural Language Processing

Recent Activity

amitkparekh updated a dataset 2 months ago

Heriot-WattUniversity/F21NL-multi30k

gpantaz authored a paper 3 months ago

Shaking Up VLMs: Comparing Transformers and Structured State Space Models for Vision & Language Modeling

gpantaz authored a paper 8 months ago

Lost in Space: Probing Fine-grained Spatial Understanding in Vision and Language Resamplers

View all activity

Heriot-WattUniversity's activity

amitkparekh

updated a dataset 2 months ago

Heriot-WattUniversity/F21NL-multi30k

Viewer • Updated Oct 21 • 30k • 1.31k

gpantaz

authored a paper 3 months ago

Shaking Up VLMs: Comparing Transformers and Structured State Space Models for Vision & Language Modeling

Paper • 2409.05395 • Published Sep 9 • 5

nateraw

posted an update 8 months ago

Post

3799

I just shared a blogpost on https://nateraw.com explaining the motivation + process of training nateraw/musicgen-songstarter-v0.2 - including training details, WandB logs, hparams, and notes on previous experiments.

Check it out here ⤵️
https://nateraw.com/posts/training_musicgen_songstarter.html

:) still kinda a WIP so if there's anything else you want to see, let me know.

3 replies

gpantaz

authored a paper 8 months ago

Lost in Space: Probing Fine-grained Spatial Understanding in Vision and Language Resamplers

Paper • 2404.13594 • Published Apr 21 • 1

nateraw

posted an update 8 months ago

Post

4248

Turns out if you do a cute little hack, you can make nateraw/musicgen-songstarter-v0.2 work on vocal inputs. 👀

Now, you can hum an idea for a song and get a music sample generated with AI 🔥🔥

Give it a try: ➡️ nateraw/singing-songstarter ⬅️

It'll take your voice and try to autotune it (because let's be real, you're no michael jackson), then pass it along to the model to condition on the melody. It works surprisingly well!