SEP 22, 2018 9:15 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 is a scientist, educator, and life-long learner with a passion for making science more communicable. When not busy in the lab isolating blood macrophages, she enjoys writing on various STEM topics.
You May Also Like
APR 01, 2021
Technology
Advancing Next-Generation Electronics
APR 01, 2021
Advancing Next-Generation Electronics
How do you envision next-gen electronics? Researchers are hoping it would be rigid-free, component-free, bendable and fl ...
APR 14, 2021
Technology
How Can Mario Kart Teach Us About Improving World Poverty?
APR 14, 2021
How Can Mario Kart Teach Us About Improving World Poverty?
Rushing down Rainbow Road and receiving power from the floating square icons on the screen, or even slipping on a banana ...
JUN 17, 2021
Infographics
Cryptocurrencies 101
JUN 17, 2021
Cryptocurrencies 101
The first cyrptocurrency, Bitcoin, was invented by an unknown person or group of people under the name Satoshi Naka ...
JUN 28, 2021
Drug Discovery & Development
Ultra-Thin Spinal Implants Could Treat Severe Pain
JUN 28, 2021
Ultra-Thin Spinal Implants Could Treat Severe Pain
Researchers from the University of Cambridge have built a minimally invasive, ultra-thin, inflatable spinal cord stimula ...
JUL 08, 2021
Earth & The Environment
New System Predicts Future Cod Supplies for Next 10 Years
JUL 08, 2021
New System Predicts Future Cod Supplies for Next 10 Years
Cod stocks in the North Sea and the Barents Sea may be predicted ten years in advance for the first time, thanks to a ne ...
JUL 13, 2021
Clinical & Molecular DX
Friend or Foe? New Cancer-Tracking Pen Guides Surgeons.
JUL 13, 2021
Friend or Foe? New Cancer-Tracking Pen Guides Surgeons.
Researchers have developed the first diagnostic “pen” that acts as a guide for surgeons, helping them distin ...
Loading Comments...