Transforming Youth Sports Video Production: SkillShow’s Automated Solution with AWS

Co-Written by Tom Koerick from SkillShow

This post explores how SkillShow utilized AWS machine learning services to revolutionize their video processing workflow in the booming youth sports market. Discover how automation reduced editing time and costs while enhancing operational efficiency.

Transforming Youth Sports Video Production with AWS: A Collaborative Approach

This post is co-written with Tom Koerick from SkillShow.

The youth sports market has seen remarkable growth, valued at a staggering $37.5 billion globally in 2022. With projections indicating a 9.2% annual growth through 2030, an estimated 60 million young athletes are actively participating worldwide. At the forefront of capturing this dynamic landscape is SkillShow, a leader in youth sports video production that films over 300 events annually, creating content for more than 20,000 young athletes.

In this post, we will explore how SkillShow leveraged Amazon Transcribe and other Amazon Web Services (AWS) machine learning services to automate their video processing workflow. The result? Reduced editing times, lowered costs, and the ability to scale their operations effectively.

The Challenge

With the surge in youth sports video production, traditional manual editing processes have become a bottleneck. Since 2001, SkillShow has stood out as a comprehensive sports video service provider, filming, editing, and distributing videos that allow athletes to showcase their skills, enhance their personal brands on social media, and support their training development.

Despite SkillShow’s market leadership and widely recognized partnerships with major organizations like Perfect Game and USA Baseball, they faced significant challenges due to their linear operations. With a lean team of just seven full-time employees, SkillShow outsourced to over 1,100 contractors annually. This reliance on outsourced editing not only inflated operational costs but also extended turnaround times to an unsustainable three weeks per event.

Managing 230 TB of video data each year compounded these challenges. Long upload times, expensive storage costs, and complex data management requirements strained their technical and IT infrastructure. With rising demand for quick content delivery—especially vital in the post-COVID era—SkillShow recognized an urgent need for an efficient, scalable solution to maintain its industry position.

Solution Overview

To tackle these challenges, SkillShow partnered with AWS to establish an automated video processing pipeline. Their team evaluated several methods for automating player identification:

• Facial Recognition: Encountered difficulties due to video quality inconsistencies and player equipment obscuring faces.

• Text-Based Detection: Initial promise waned as jersey numbers became obscured with player movements and environmental conditions.

Ultimately, the team opted for an audio logging and automated clip generation solution, which provided several key advantages:

• Reliable Identification: Announcers consistently calling out player details proved more dependable.

• Environmental Resilience: Audio quality remained stable, ensuring better detection during challenging conditions.

• Cost-Effectiveness: Lesser computational demands simplified processing and increased accuracy rates.

This solution capitalized on several key AWS services:

Amazon S3

• Serves as scalable, durable storage for both input and output video files, accommodating SkillShow’s massive 230 TB annual data volume.

AWS Lambda

• Enables event-driven processing, seamlessly automating tasks like transcription and clip generation without infrastructure management.

Amazon Transcribe

• Converts video audio into accurate text transcripts, allowing deeper analysis and player identification, even under noisy conditions.

The architectural flow of the solution is as follows:

1. An authorized user uploads a .csv roster and video footage.
2. A Lambda function triggers upon the video upload.
3. Amazon Transcribe generates a timestamped transcript.
4. The transcript is processed by another Lambda function to segment the video based on roster information.

By integrating these AWS services, SkillShow built a scalable, cost-effective video processing solution that resolves operational hurdles while accommodating their growing data demands.

Example Processing Workflow

Imagine this streamlined workflow:

1. Upload a player roster .csv and video file to the input bucket.
2. The auto-transcribe function processes the audio.
3. The auto-clipper function segments the video based on player information.
4. Final clips are organized into designated output folders, streamlining editor access.

This method allows editors to focus on reviewing files that require manual renaming, enhancing efficiency exponentially.

Results and Benefits

After implementing the AWS-powered solution, SkillShow drastically transformed its operations. The automated pipeline slashed video production time from three weeks to just 24 hours per event. In a recent Chicago event, the system processed 69 clips, accurately editing 64—a 93% success rate. This high accuracy proved its capability to manage real-world scenarios effectively.

During the Northwest Indoor event, the automated process managed approximately 270 clips with over 90% accuracy while handling diverse footage types, showcasing the solution’s adaptability.

With this new workflow, SkillShow can process multiple events concurrently, significantly enhancing its service delivery to youth sports leagues. These improvements not only facilitated faster content delivery but also elevated the viewing experience for players, coaches, and scouts alike.

Conclusion

SkillShow’s journey highlights how AWS ML services can revolutionize resource-intensive video processing workflows into seamless, automated systems. The combination of scalable storage with Amazon S3, serverless computing via Lambda, and the speech recognition capabilities of Amazon Transcribe set a new standard for operational efficiency.

This innovative approach not only serves the youth sports market but also opens the door for additional applications across various industries.

Interested in enhancing your video processing capabilities? Contact SkillShow to discover how their AWS-powered pipeline can transform your content production process.

About the Authors

Ragib Ahsan is a Partner Solutions Architect at Amazon Web Services (AWS), specializing in helping organizations implement AI/ML solutions. With a focus on serverless architecture, he champions creating practical applications with cloud technologies.

Tom Koerick is the CEO of SkillShow, a sports media network firm dedicated to filming youth sporting events nationwide. Formerly a professional baseball player, Tom now develops video solutions that help athletes and organizations thrive in the youth sports arena.