SEP 22, 2018 09:15 PM PDT

Tackling Speech and Object Recognition

WRITTEN BY: Nouran Amin

Image via Tech Hive

Computer scientists at MIT have opened new doors in speech and image recognition systems by creating a model system that can identify objects within an image based solely on the spoken descriptions of the image; an audio caption. Despite current speech-recognition technology, the new model will not need manual transcriptions and annotations of the examples it is trained on. The system instead adapts to words directly from recorded speech clips and objects placed raw images, and then associates them with one another. Even though the new systems currently recognizes only several different hundred words, researchers are hopeful in the future that their combined speech-object recognition technique can be useful instead of hours of manual labor

"We wanted to do speech recognition in a way that's more natural, leveraging additional signals and information that humans have the benefit of using, but that machine learning algorithms don't typically have access to. We got the idea of training a model in a manner similar to walking a child through the world and narrating what you're seeing," says David Harwath, a researcher in the Computer Science and Artificial Intelligence Laboratory (CSAIL) and the Spoken Language Systems Group.

Additionally, on useful application of the new system is replacing a bilingual annotator by learning the translations between languages, without the need of a bilingual annotator. "There's potential there for a Babel Fish-type of mechanism," explains Harwath by referring to the fictitious living earpiece in the "Hitchhiker's Guide to the Galaxy".

Image via Electronic Design

In their research paper, the scientists altered the model system to combine specific words with patches of pixels. They trained the model on a database system giving correct and incorrect images with captions. However, there exists a challenge during the training where the model doesn’t have access to alignment information between the speech and the image. "The biggest contribution of the paper," Harwath explains, "is demonstrating that these cross-modal [audio and visual] alignments can be inferred automatically by simply teaching the network which images and captions belong together and which pairs don't."

Source: MIT news

About the Author
  • Nouran enjoys writing on various topics including science & medicine, global health, and conservation biology. She hopes through her writing she can make science more engaging and communicable to the general public.
You May Also Like
OCT 23, 2019
Chemistry & Physics
OCT 23, 2019
New Battery Tech "Squeezes" More Electricity out of Seawater
Oceans make up 71% of the Earth's surface, and it also harbors one of the world's greatest yet large untouched energy source: osmotic power. The sa...
OCT 23, 2019
Technology
OCT 23, 2019
Can Robots Land Like Birds?
Below the gaze of high-speed cameras, a tiny bird named Gary is awaiting the signal to fly over to a perch covered in Teflon. The successful land of Gary o...
OCT 23, 2019
Chemistry & Physics
OCT 23, 2019
Irish Teen Removes Microplastics From Water, Wins Google Science Fair
A new invention that removes microplastics from water was designed by 18-year-old Fionn Ferreira -- winner of the Google Science Fair....
OCT 23, 2019
Technology
OCT 23, 2019
Artificial Muscle for Soft Robotics
Imagine wearing a flower brooch that blooms in front of your eyes? Well, the ultrathin, artificial muscle used for soft robotics was recently developed by ...
OCT 23, 2019
Space & Astronomy
OCT 23, 2019
What We've Learned About Mars From India's Mangalyaan Mission
Back in 2013, the Indian Space Research Organization (ISRO) launched a Mars-orbiting spacecraft dubbed the Mars Orbiter Mission (MOM), also known as Mangal...
OCT 23, 2019
Technology
OCT 23, 2019
Telescope technology takes first accurate images of glaucoma-related eye structure
Using the same tools designed to observe the stars, vision scientists at Indiana University have taken the first accurate microscopic images of the trabecu...
Loading Comments...