Virtual Race Audio Commentary Generation

Generated commentary combined with externally created visuals

Highlights

• Replaced the previous outsourced commentary solution that had varied quality and required extensive QA.
• Algorithmically generated race commentary by combining race meta data of runner positions with pre-recorded audio clips of runner numbers, positions and phrases.
• Built the replacement from scratch in short timescales, resulting in commentary of reliably higher quality, increased flexibility, diagnostic reports and far less QA effort required.
• Implementation was applied to the data and content for generation of horse racing and greyhound racing commentary.

Responsibilities

• Analysis, design, development, unit testing.
• Lead developer of whole solution.

Solution

Technologies: Java, Eclipse, Windows, macOS, ffmpeg

Achievements

• Provided increased usage of the recorded script phrases than the previous outsourced solution, improving the variation of commentary generated.
• Reduced production turnaround of phrase and script changes from two weeks to several minutes.
• Reduced QA rejection of race commentary issues from 30% to less than 5%.

Generated commentary combined with externally created visuals