136
MatAnyone
🤡
Gradio demo for MatAnyone
Testing
Python2231
·
AI & ML interests
None yet
Recent Activity
liked
a Space
16 days ago
PeiqingYang/MatAnyone
liked
a Space
20 days ago
opencompass/open_vlm_leaderboard
liked
a dataset
21 days ago
homebrewltd/Maze-Reasoning-v0.1
Organizations
None yet
Python2231's activity
liked
a
Space
16 days ago
liked
a
Space
20 days ago
liked
a
dataset
21 days ago
liked
a
dataset
22 days ago
reacted to
smirki's
post with 🔥
22 days ago
Post
3381
UIGEN for Tailwind v4 is coming soon!
- 2 replies
liked
a
Space
22 days ago
liked
a
Space
25 days ago
liked
a
Space
28 days ago
liked
a
dataset
about 1 month ago
reacted to
s-emanuilov's
post with 🔥
about 1 month ago
Post
5193
Tutorial 💥 Training a non-English reasoning model with GRPO and Unsloth
I wanted to share my experiment with training reasoning models in languages other than English/Chinese.
Using Llama 3.1 8B as base, GRPO trainer from trl, and Unsloth optimizations, I got a working prototype in Bulgarian after ~5 hours on an L40S GPU. The approach should work for any language where the base model has some pre-training coverage.
Full code and tutorial here: https://unfoldai.com/reasoning-in-a-non-english-language/
The model itself: s-emanuilov/LLMBG-Llama-3.1-8B-BG-Reasoning-v0.1
I hope this helps anyone looking to build reasoning models in their language.
I wanted to share my experiment with training reasoning models in languages other than English/Chinese.
Using Llama 3.1 8B as base, GRPO trainer from trl, and Unsloth optimizations, I got a working prototype in Bulgarian after ~5 hours on an L40S GPU. The approach should work for any language where the base model has some pre-training coverage.
Full code and tutorial here: https://unfoldai.com/reasoning-in-a-non-english-language/
The model itself: s-emanuilov/LLMBG-Llama-3.1-8B-BG-Reasoning-v0.1
I hope this helps anyone looking to build reasoning models in their language.
- 4 replies
liked
a
Space
about 1 month ago
reacted to
hexgrad's
post with 👍
about 1 month ago
Post
5665
I wrote an article about G2P: https://hf.co/blog/hexgrad/g2p
G2P is an underrated piece of small TTS models, like offensive linemen who do a bunch of work and get no credit.
Instead of relying on explicit G2P, larger speech models implicitly learn this task by eating many thousands of hours of audio data. They often use a 500M+ parameter LLM at the front to predict latent audio tokens over a learned codebook, then decode these tokens into audio.
Kokoro instead relies on G2P preprocessing, is 82M parameters, and thus needs less audio to learn. Because of this, we can cherrypick high fidelity audio for training data, and deliver solid speech for those voices. In turn, this excellent audio quality & lack of background noise helps explain why Kokoro is very competitive in single-voice TTS Arenas.
G2P is an underrated piece of small TTS models, like offensive linemen who do a bunch of work and get no credit.
Instead of relying on explicit G2P, larger speech models implicitly learn this task by eating many thousands of hours of audio data. They often use a 500M+ parameter LLM at the front to predict latent audio tokens over a learned codebook, then decode these tokens into audio.
Kokoro instead relies on G2P preprocessing, is 82M parameters, and thus needs less audio to learn. Because of this, we can cherrypick high fidelity audio for training data, and deliver solid speech for those voices. In turn, this excellent audio quality & lack of background noise helps explain why Kokoro is very competitive in single-voice TTS Arenas.
liked
a
Space
about 1 month ago
reacted to
victor's
post with ❤️
about 1 month ago
Post
4210
Hey everyone, we've given https://hf.co/spaces page a fresh update!
Smart Search: Now just type what you want to do—like "make a viral meme" or "generate music"—and our search gets it.
New Categories: Check out the cool new filter bar with icons to help you pick a category fast.
Redesigned Space Cards: Reworked a bit to really show off the app descriptions, so you know what each Space does at a glance.
Random Prompt: Need ideas? Hit the dice button for a burst of inspiration.
We’d love to hear what you think—drop us some feedback plz!
Smart Search: Now just type what you want to do—like "make a viral meme" or "generate music"—and our search gets it.
New Categories: Check out the cool new filter bar with icons to help you pick a category fast.
Redesigned Space Cards: Reworked a bit to really show off the app descriptions, so you know what each Space does at a glance.
Random Prompt: Need ideas? Hit the dice button for a burst of inspiration.
We’d love to hear what you think—drop us some feedback plz!
- 5 replies
upvoted
an
article
about 1 month ago
Article
DEMO: French Spoken Language Understanding with the new speech resources from NAVER LABS Europe
By
•
•
10
liked
4
Spaces
about 1 month ago
2
Speechbrain Lang Id Voxlingua107 Ecapa
💬
Identify language in audio recordings
13
Whisper Turbo Subtitle
🚀
Generate Subtitle Using faster-whisper-large-v3-turbo-ct2
250
Kokoro TTS Zero
🎴
✨[With v1.0.0] Accelerated TTS on Kokoro-82M
198
ClearerVoice-Studio (Speech Enhancement, Separation and Extraction)
📈
Better AI powered platform to purify your speech signal
reacted to
alibabasglab's
post with 👍
about 2 months ago
Post
2344
Do you need to improve your speech audio to premium quality? If so, please try out our latest open-sourced free speech processing toolkit: [ClearerVoice-Studio](https://github.com/modelscope/ClearerVoice-Studio)! Check out our live demo at
alibabasglab/ClearVoice
and https://modelscope.cn/studios/iic/ClearerVoice-Studio.
and https://modelscope.cn/studios/iic/ClearerVoice-Studio.
- 1 reply