We’re judging each submission based on four key criteria, each worth 25% of the overall submission score. Each criterion will be judged by one or more experts:

Criterion	Assessment
Voice similarity: how similar do the submission voices sound to the Agent Luke and Chris Nemesis reference samples?	Wespeaker: Automatic speaker recognition measure – 12.5% Dr Georgina Brown: Forensic Speech Scientist – 12.5%
Naturalness: how natural, spontaneous, and human-like is the speech and conversation?	Dr Justin Lo: Forensic Speech Scientist – 12.5% Prof Beatrice Szczepek-Reed: Conversation Analyst – 12.5%
Content: does the submission contain (a) Chris urging the door guards to let Luke in, and (b) Luke referencing a childhood memory supposedly shared with Chris?	Dr Lena Podoletz: Criminologist – 12.5% Prof Claire Hardaker: Forensic Linguist – 12.5%
Creative flair: to what extent does the submission show creative flair and careful attention to detail in enhancing the realism of the interaction?	Dr Georgina Brown: Forensic Speech Scientist – 5% Dr Justin Lo: Forensic Speech Scientist – 5% Dr Lena Podoletz: Criminologist – 5% Prof Beatrice Szczepek-Reed: Conversation Analyst – 5% Prof Claire Hardaker: Forensic Linguist – 5%

How do you know your submissions will be judged by experts? Read more about The Panel next.