Hertz-vae: > 1.8B parameters, 8-layer decoder-only transformer
> First four layers receive latent history
> Layer 5 receives ground-truth 15-bit quantized representation during training
> Directly samples hertz-lm's next token prediction during inference
> Near-perfect at
SOFTWARE
-
Hertz-VAE: 1.8B Parameter Decoder-Only Transformer Architecture
By
–
-
Hertz-codec: Advanced Convolutional Audio VAE Outperforms Competitors
By
–
Hertz-codec: > Convolutional audio VAE
> Encodes 16kHz mono speech to 8Hz latent representation at 1kbps
> 32-dim latent per 125ms frame
> Outperforms Soundstream and Encodec at 6kbps, on par with DAC at 8kbps
> 5M encoder, 95M decoder parameters -
GitHub repository link for Voice Pro project
By
–
github: https://
github.com/abus-aikorea/v
oice-pro
… -
Voice-Pro: Gradio Interface for Audio Transcription, Translation, and Speech
By
–
Voice-Pro
— AK (@_akhaliq) 9 novembre 2024
gradio web-ui for transcription, translation and text-to-speech pic.twitter.com/Fr2rWfKGoJVoice-Pro gradio web-ui for transcription, translation and text-to-speech
-
Cost-Optimizing Chatbots with Semantic Caching on Databricks
By
–
As businesses prepare to widely deploy chatbots into production, many are encountering a significant challenge: cost Learn how Databricks provides an optimal platform for building cost-optimized chatbots with semantic caching capabilities. https://
bit.ly/3YptydL -
X Pro outperforms phone version without refreshing
By
–
You miss X Pro. It runs without refreshing so it’s way better than phone
-
Tesla Autonomous Vehicle Powers Silicon Valley Daily Commutes
By
–
My Tesla is in my garage and drives me everywhere in Silicon Valley. And the world.
-
Google Sr Manager Enhances Software Engineer Productivity
By
–
🔦✨ Shining a spotlight on Maxim Tabachnyk, Sr Engineering Mgr at Google. Maxim works to enhance software engineer productivity across the entire development lifecycle, including writing, reviewing, & maintaining code. Hear from him + check out the blog ↓ https://t.co/VuWmr3FNyN pic.twitter.com/aqdiDDMvAJ
— Google AI (@GoogleAI) 8 novembre 2024Shining a spotlight on Maxim Tabachnyk, Sr Engineering Mgr at Google. Maxim works to enhance software engineer productivity across the entire development lifecycle, including writing, reviewing, & maintaining code. Hear from him + check out the blog ↓
-
Real-Time Object Detection YOLOv7 Mobile Phone Implementation
By
–
New video! Detecting Objects in REAL-TIME with YOLOv7 on Your Phone! Learn more: https://
youtu.be/Dd8cQ2GZPRY #yolo #yolov7 #computervision #ai -
SemiAnalysis joins Passport enterprise software for creators
By
–
Great to see @SemiAnalysis_ on Passport. Enterprise-level software for creators