48 50 58

Steven Zheng

Steveeeeeeen

AI & ML interests

speech & audio

Recent Activity

liked a Space about 14 hours ago

ACE-Step/ACE-Step

upvoted a paper about 14 hours ago

LLaMA-Omni2: LLM-based Real-time Spoken Chatbot with Autoregressive Streaming Speech Synthesis

new activity about 14 hours ago

naver-hyperclovax/HyperCLOVAX-SEED-Vision-Instruct-3B:Update library tag for better download tracking and code snippets!

View all activity

Organizations

Steveeeeeeen's activity

liked a Space about 14 hours ago

ACE Step

😻

A Step Towards Music Generation Foundation Model

upvoted a paper about 14 hours ago

LLaMA-Omni2: LLM-based Real-time Spoken Chatbot with Autoregressive Streaming Speech Synthesis

Paper • 2505.02625 • Published 1 day ago • 10

New activity in naver-hyperclovax/HyperCLOVAX-SEED-Vision-Instruct-3B about 14 hours ago

Update library tag for better download tracking and code snippets!

#16 opened about 14 hours ago by

Steveeeeeeen

New activity in deepseek-ai/DeepSeek-Prover-V2-7B about 14 hours ago

Add model card metadata

#1 opened about 14 hours ago by

Steveeeeeeen

New activity in SeaLLMs/SeaLLMs-Audio-7B about 17 hours ago

Update library tag for better download tracking and code snippets!

#1 opened about 17 hours ago by

Steveeeeeeen

New activity in moonshotai/Kimi-Audio-7B about 17 hours ago

Update library tag for better download tracking and code snippets!

#2 opened about 17 hours ago by

Steveeeeeeen

Update library tag for better download tracking and code snippets!

#1 opened about 17 hours ago by

Steveeeeeeen

New activity in moonshotai/Kimi-Audio-7B-Instruct about 17 hours ago

Update library tag for better download tracking and code snippets!

#13 opened about 17 hours ago by

Steveeeeeeen

upvoted a paper about 18 hours ago

Voila: Voice-Language Foundation Models for Real-Time Autonomous Interaction and Voice Role-Play

Paper • 2505.02707 • Published 1 day ago • 56

reacted to fdaudens's post with 👍🔥 2 days ago

Post

2848

Forget everything you know about transcription models - NVIDIA's parakeet-tdt-0.6b-v2 changed the game for me!

Just tested it with Steve Jobs' Stanford speech and was speechless (pun intended). The video isn’t sped up.

3 things that floored me:
- Transcription took just 10 seconds for a 15-min file
- Got a CSV with perfect timestamps, punctuation & capitalization
- Stunning accuracy (correctly captured "Reed College" and other specifics)

NVIDIA also released a demo where you can click any transcribed segment to play it instantly.

The improvement is significant: number 1 on the ASR Leaderboard, 6% error rate (best in class) with complete commercial freedom (cc-by-4.0 license).

Time to update those Whisper pipelines! H/t @Steveeeeeeen for the finding!

Model: nvidia/parakeet-tdt-0.6b-v2
Demo: nvidia/parakeet-tdt-0.6b-v2
ASR Leaderboard: hf-audio/open_asr_leaderboard