35 40 95

Eric Bezzam PRO

bezzam

AI & ML interests

speech, audio, imaging

Recent Activity

upvoted an article about 3 hours ago

Did GPT 5.2 make a breakthrough discovery in theoretical physics?

updated a model 2 days ago

bezzam/VibeVoice-ASR-7B

updated a model 2 days ago

bezzam/VibeVoice-ASR-7Bv2

View all activity

Organizations

upvoted an article about 3 hours ago

Article

Did GPT 5.2 make a breakthrough discovery in theoretical physics?

about 18 hours ago

•

upvoted an article 3 days ago

Article

Compute and Competition in AI: Different FlOPs for Different Folks

8 days ago

•

upvoted an article 4 days ago

Article

How to Build a Benchmark with a Private Test Set on Hugging Face

4 days ago

•

upvoted a collection 14 days ago

OWSM: Fully Open Speech Recognition and Translation Models

Collection

A collection of models related to the Open Whisper-style Speech Models (OWSM) project from CMU: https://www.wavlab.org/activities/2024/owsm/ • 21 items • Updated Mar 8, 2025 • 3

upvoted an article 15 days ago

Article

Community Evals: Because we're done trusting black-box leaderboards over the community

17 days ago

•

upvoted 3 articles 17 days ago

Article

The Future of the Global Open-Source AI Ecosystem: From DeepSeek to AI+

17 days ago

•

Article

Architectural Choices in China's Open-Source AI Ecosystem: Building Beyond DeepSeek

24 days ago

•

Article

One Year Since the “DeepSeek Moment”

about 1 month ago

•

upvoted a collection 19 days ago

Treble10

Collection

Treble Technologies and Hugging Face have entered in to a long term collaboration. In celebration, we are releasing the Treble10 dataset. • 3 items • Updated Oct 28, 2025 • 5

upvoted a paper 24 days ago

VIBEVOICE-ASR Technical Report

Paper • 2601.18184 • Published 25 days ago • 20

upvoted 2 articles about 1 month ago

Article

Open Responses: What you need to know

Jan 15

•

107

Article

Scaling Real-Time Voice Agents with Cache-Aware Streaming ASR

Jan 5

•

upvoted an article about 2 months ago

Article

Blazingly fast whisper transcriptions with Inference Endpoints

May 13, 2025

•

upvoted 2 papers 2 months ago

Voxtral

Paper • 2507.13264 • Published Jul 17, 2025 • 32

Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition

Paper • 2305.05084 • Published May 8, 2023 • 4

upvoted 2 articles 2 months ago

Article

Shadow AI - Where are the CIOs?

Dec 19, 2025

•

Article

Tokenization in Transformers v5: Simpler, Clearer, and More Modular

Dec 18, 2025

•

120

upvoted 3 articles 3 months ago

Article

We Got Claude to Fine-Tune an Open Source LLM

Dec 4, 2025

•

596

Article

Transformers v5: Simple model definitions powering the AI ecosystem

Dec 1, 2025

•

298

Article

Continuous batching from first principles

Nov 25, 2025

•

326

Eric Bezzam PRO

AI & ML interests

Recent Activity

Organizations

bezzam's activity

Did GPT 5.2 make a breakthrough discovery in theoretical physics?

Compute and Competition in AI: Different FlOPs for Different Folks

How to Build a Benchmark with a Private Test Set on Hugging Face

Community Evals: Because we're done trusting black-box leaderboards over the community

The Future of the Global Open-Source AI Ecosystem: From DeepSeek to AI+

Architectural Choices in China's Open-Source AI Ecosystem: Building Beyond DeepSeek

One Year Since the “DeepSeek Moment”

Open Responses: What you need to know

Scaling Real-Time Voice Agents with Cache-Aware Streaming ASR

Blazingly fast whisper transcriptions with Inference Endpoints

Shadow AI - Where are the CIOs?

Tokenization in Transformers v5: Simpler, Clearer, and More Modular

We Got Claude to Fine-Tune an Open Source LLM

Transformers v5: Simple model definitions powering the AI ecosystem

Continuous batching from first principles