cure.fit Receives AWS Support for Health and Fitness Live Streams
cure.fit is a health and fitness company based in India that offers a holistic range of services including fitness and yoga, healthy meals, mental wellbeing, and primary care. Based in India, The company cure.fit operates more than 200 fitness centers and the offers services for at-home health and fitness are also accessible via its cult.live app for at home health and fitness. With the aim to make working out fun and easy, cult.live app delivers live-streaming classes enhanced with deep -learning, computer vision, and gamification.
The configurability of AWS’ services and their responsive customer support make it simple to experiment and design the optimal experience for cure.fit’s users. Together with the auto-management and robust scaling capabilities of the AWS Media Services and Amazon CloudFront, we can rely on AWS for best-in-class video streaming, analytics and machine learning while focusing on cure.fit’s business.”
“Prior to lockdown in India, we were planning to launch an app so that fitness fans in locations that did not have a cure.fit fitness center could join cure.fitour exercise classes. We fast-tracked the cult.live app for a March 2020 launch so that our users could continue to maintain their fitness and enjoy the experience of our classes while exercising at home,” says Kunal Chourasia, Head of Engineering, cure.fit. “As health and fitness lifestyles continues to evolve post-pandemic, we have increased the number of live classes and feedback for users via the app.”
When the COVID-19 pandemic hit India and the lockdown was enforced, the traffic on cure.fit live classes spiked 30 times the usual volume within a week. Within a span of three weeks, the app was re-architected to enable nearly 300,000 users to attend a class at the same time.
Today the cult.fit app is features more than 500 streaming workout sessions per day, and more than half a million workout sessions are viewed live and on-demand per day. Users can tune in to classes on their mobile device, desktop or connected TVs.
Live Video and Analytics Workflow
During the cult.live exercises live classes users receive real-time feedback from the fitness instructors, who are trainers and experienced athletes. The live video is delivered with minimal buffering over varying network conditions, including cellular.
cure.fit Streaming Workflow
Cameras capturing celebrity instructors feed to an in-house content management system that stores the video in Amazon Simple Storage Service (Amazon S3).
Amazon S3 queues AWS Elemental MediaLive and AWS Elemental MediaConvert to encode the video into three bitrate tiers, packaged using AWS Elemental MediaPackage. The tiered bitrates ensure a smooth viewing experience even if network instabilities occur.
Amazon CloudFront then delivers content in an optimized, scalable manner. When a class starts, the viewing traffic spikes. Users follow the instructor and compete against each other, so only a few seconds of latency is acceptable.
Amazon Elastic Cloud Compute (Amazon EC2) P3 instances train machine learning models that run on mobile devices. The analytics workflow and feedback loop enables detailed app and network analysis.
The cult.live app’s Energy Meter acts as a virtual movement tracker, and is an important element to giving users an engaging experience. It utilizes machine learning to understand how the user moves during the workout, how accurately they are performing the exercises, and approximately how much energy they are burning while moving a particular body part during the workout. The Energy Meter gives users real-time feedback about their workout, and shows stats to in compete comparison with others working out in the same session.
“The configurability of AWS’ services and their responsive customer support make it simple to experiment and design the optimal experience for cure.fit’s users. Together with the auto-management and robust scaling capabilities of the AWS Media Services and Amazon CloudFront, we can rely on AWS for best-in-class video streaming, analytics, and machine learning while focusing on cure.fit’s business,” says Chourasia.