Oct 7, 2020

Machine learning for music: Google’s Tone Transfer

AI
Machine Learning
music
Google
Paddy Smith
2 min
google magenta tone transfer
Google’s DDSP – better known as Tone Transfer – is a tool for making music. But will it help organise voice data in the future...

Google made a low-key launch this week, when it made Tone Transfer public. Built by two teams at Mountain View – Magenta and AI UX – the tool takes a tonal input (a voice or a line of melody) and can then re-render it with instrument modelling.

The year-long collaboration between AI researchers, UX engineers and designers is built on Magenta’s Differential Digital Signal Processing engine (DDSP). It was created as an exercise in learning about how people perceive music, machine learning and their own practice. Running on an early version of DDSP, it’s an in-browser deployment (on tensorflow.js) which extracts pitch data using another Google Research project, SPICE .

At the moment, the tool has been opened up for experiment by musicians and non-musicians who want to explore music creation. But Magenta made DDSP open source earlier this year and, while it hasn’t been expressly stated, there may be implications for business data collection.

As new routes to mining data are explored, and voice collection via VOIP services becomes more commonplace, there is scope to explore tonal approaches to voice data. Applying machine learning to voice data could help to parse language that otherwise could be misinterpreted by data analysts. Sarcasm or humour are both common idioms that would create a false positive if voice data is mined in like-for-like fashion with text.

Magenta says: “We are excited with upcoming releases enabling you to easily train your own DDSP models and deploy them everywhere: a phone, an audio plugin or a website using the larger tensorflow lite and tensorflow.js ecosystem.”

Data scientists, always on the lookout for a new frontier, should prick their ears.

Share article

Jul 28, 2021

AI pioneer ZAC named as a Top 5 Leading Global Company

ZAC
Cognitive Explainable-AI
AI
Image recognition
Catherine Gray
2 min
Z Advanced Computing (ZAC) has been recognised as one of the Top 5 Leading Global Companies in The Fourth Industrial Revolution by an Oxford Academic paper

The pioneer Cognitive Explainable-Artificial Intelligence (Cognitive XAI) start-up, ZAC, offers a disruptive 3D image recognition and search platform based on Explainable-AI. Its technology, which ZAC claims beats the competition ‘hands down’, has gained recognition from an Oxford Academic paper naming it one of the Top 5 Leading Global Companies in The Fourth Industrial Revolution.

ZAC has made AI and Machine Learning (ML) breakthroughs. Using only a few training samples and an average laptop with low power CPU, ZAC has achieved complex Image Recognition.

This is a sharp contrast to the other algorithms in the industry that require thousands to billions of training samples, trained on large GPU servers.

"ZAC requires much less computing power and much less electrical power to run, which is great for mobile and edge computing, as well as the environment, with less Carbon footprint,” said Dr. Bijan Tadayon, CEO and co-founder of ZAC.

“You cannot do this with the other algorithms, such as Deep Convolutional Neural Networks (CNN) or ResNets, even with an extremely large number of training samples, on GPU servers," Tadayon explained.

Removing barriers in image recognition

Founded by three siblings, ZAC’s platform overcomes a major limitation in image recognition of consumer products. 

Previous technology fails to recognise details beyond generic categories or classifications as existing AI technologies only produce a generic level of output.

With its platform ZAC “are removing today’s barriers and limitations in image recognition... We have figured out how to apply Explainable-AI to machine learning for the first time in the industry,” said Tadayon.

AI breakthrough for a variety of applications

ZAC’s revolutionary AI breakthrough in 3D image recognition mimics how humans discover, recognise and learn. It is also able to detect fine details in images from all views and angles.

Its horizontal platform features tools and APIs to enable a wide variety of applications. This means that multiple industries can take advantage of the platform’s efficient, detailed and accurate recognition and search.

The platform can be used in a variety of applications, some of which include:

  • Medical imaging and diagnosis
  • E-commerce and retail
  • Satellite and aerial image analysis
  • Facial recognition
  • Imagery detection and analysis

By providing a cognitive-based image recognition that is fully scalable and faster than current technology, it is clear to see why ZAC has been named one of the Top 5 Leading Global Companies in the Fourth Industrial Revolution

Share article