Reducing speech to text model training time on switchboard 2000 from a week to under two hours – Artofit

Image gallery for: Reducing speech to text model training time on switchboard 2000 from a week to under two hours

Reducing Speech-to-Text Model Training Time on Switchboard-2000 from a Week to Under Two Hours

In our ICASSP 2020 paper, we successfully shorten the training time on the 2000-hour Switchboard dataset from over a week to less than two hours.
Advertisement
Motion Graphics Inspiration - @joyce.n.ho

Thing cloud
Portfolio
Creative code
Quick Saves
Advertisement
5.2K views · 2.2K reactions | Another triangle surprise 😮 #mathematics #maths #math #geometry | Tim Brzezinski | Facebook

teaching in hschool
D4RL: Building Better Benchmarks for Offline Reinforcement Learning

Deep Learning
Introducing BVR 'Bridging Visual Representations': A Novel Module And Applied Plug-In Designed To Better Integrate Different Computer Vision (CV) Object Representations

Data Science
IBM Research AI Advances Speaker Diarization in Real Use Cases

Artificial Intelligence
Adversarial Robustness Toolbox: One Year Later with v1.4

Artificial Intelligence
Layerwise learning for Quantum Neural Networks

Artificial Intelligence
How the brain produces a rhythm discovered almost 100 years ago

Artificial Intelligence