Google AI is focused on bringing the benefits of AI to everyone. In conducting and applying our research, we advance the state-of-the-art in many domains.
Today on the blog, read all about how automatically repairing non-building code increases productivity and appears to introduce no detectable negative impact on code safety, provided that high quality training data and responsible monitoring are employed →
Today we describe how federated analytics, differential privacy, and our partnerships with linguists enabled our keyboard application, Gboard, to improve its vocabulary, boosting Gboard’s performance while keeping your typed words private. Learn more at:
Stop by the
#ICASSP2024
Google booth today during the morning break (10:20 AM) where John Apostolopoulos will be hosting a Q&A on the Future of Video Communication in the Age of Gen AI.
At
#ICASSP2024
? Join us in the main auditorium (3F) at 3:10pm today, where Googler Johan Schalkwyk will be giving the afternoon Plenary Talk on "Multi-modal large language models as the path towards language inclusivity"
Miss yesterday's demonstration of our new
#ML
model for real-time multi-mic audio separation? The team will be back with a hands-on demo at the
#ICASSP2024
Google booth at 10:20 AM. Stop in and check it out!
Distinguishing speech in noisy environments can be challenging for many. Today we illustrate how training a joint sound separation + ASR model can benefit from hybrid datasets that have large amounts of simulated audio & small amounts of real recordings.
The ScaNN vector search library was open-sourced in 2020 to highlight innovations in vector search algorithms, critical for many
#ML
applications. Today, learn how SOAR introduces redundancy to ScaNN’s vector index to improve vector search efficiency →
Today at 4:10pm, stop by the
#ICASSP2024
Google booth, where Yang Yang, Shao-Fu Shih & George Sung will demo their
#ML
models for real-time multi-mic audio separation that's able to isolate target audio & speech from different angles and is adaptable to different mic geometries.
Join us today for the
#ICASSP2024
Innovation Forum Panel on "GenAI: Challenges and Opportunities for Signal Processing", moderated by Googler John Apostolopoulos. We'll see you there!
During the
#ICASSP2024
morning break today, Yang Yang, Yury Kartynnik, Shao-Fu Shih, and George Sung will be at the Google booth to demonstrate a new tool for real-time on-device voice conversion on the Pixel 8 Pro. Stop by to try the demo yourself!
The minimum cut problem is a structural Q about the connectivity of a graph & one of the key problems in algorithmic graph theory, w/ many practical applications. Read about the 1st deterministic, nearly-linear algorithm that works for general graphs →
Project Relate is a beta Android app that uses
#ML
to help people with non-standard speech communicate more easily. Today at 4:10pm, drop by the
#ICASSP2024
Google Booth (P7) where Katrin Tomanek will talk all about it. Read more ↓
Congratulations to Shrikanth Narayanan for winning the IEEE Signal Processing Society's Claude Shannon-Harry Nyquist Award, in recognition of his contributions to spoken language processing technologies and their societal applications!
#ICASSP2024
Congratulations to Hossein Talebi & Peyman Milanfar for winning the IEEE Signal Processing Society's Best Paper Award for their paper titled "NIMA: Neural Image Assessment"!
#ICASSP2024
Attending the
@ieeeICASSP
conference? Stop by Google's booth (P7) where you can chat with researchers, attend a demo session, and learn about some of our many groundbreaking research projects being presented at
#ICASSP2024
. Learn more →
The 2024
@ieeeICASSP
meeting is in full swing! Curious what Google researchers are doing in the fields of speech, acoustics, and signal processing? Learn about our many research projects, talks, papers &
#ICASSP2024
Google booth activities →