Recently, Amazon announced the general availability of Deepcomposer, a service in AWS, which provides developers with a creative way to learn Machine Learning (ML). Deepcomposer is a machine learning-enabled keyboard for developers, and available for purchase.
At re:Invent last year, Amazon launched AWS Deepcomposer as a preview, and it is generally available now with a few new features. These features include:
- Learning Capsules: providing developers with tutorials to learn Generative AI in easy-to-consume, bite-sized modules.
- In-Console Training: enabling developers to train their generative models in the AWS DeepComposer console, without having to write a single line of machine learning code.
- Rhythm Assist: aligns musical notes users play on the keyboard to the closest beat.
Deepcomposer belongs to the same family of AWS hardware as DeepLens and DeepRacer with focus on teaching developers specific machine learning capabilities. With Deepcomposer, the goal is to allow developers to learn about generative adversarial networks (GAN), a neural network architecture explicitly built to generate new samples from an existing data set.
Julien Simon, artificial intelligence & machine learning evangelist for EMEA, states in a blog post on the GA of Deepcomposer:
Until now, developers interested in growing skills in GANs haven’t had an easy way to get started. In order to help them regardless of their background in ML or music, we are building a collection of easy learning capsules that introduce key concepts, and how to train and evaluate GANs. This includes a hands-on lab with step-by-step instructions and code to build a GAN model.
Source: https://aws.amazon.com/blogs/aws/aws-deepcomposer-now-generally-available-with-new-features/
A GAN puts different neural networks against each other to produce original digital works based on sample inputs, for instance, from the Deepcomposer keyboard. With Deepcomposer service, users can train and optimize GAN models to create original music. For example, a user plays a short melody using the keyboard or on-screen one, and the service then automatically generates a backing track based on your choice of musical style.
Gillian Armstrong, a solution architect at Liberty IT, wrote in a medium.com post:
The DeepComposer Studio has a range of ways of interacting with it, so people of all levels (and people moving through levels) can get to work right away. DeepComposer needs two inputs to work:
- A short piece of music
- A machine learning model (this is what will be used to create the composition)
Lastly, customers can now try Deepcomposer out on an AWS free tier and buy the keyboard for $99 at Amazon.com. Moreover, the keyboard is currently only available in the US with access to the AWS Deepcomposer Console from the US East region. Note that AWS customers can also use Deepcomposer without the keyboard. Further pricing details for consuming the service is available on the pricing page.
Credit: Google News
