Understand your results#

This section contains information about the different types of results you get from Safe Voice.

Safe Voice uploads the following results to the Moderation Platform:

• Audio recordings: Recorded audio from the moment a player submitted a report
• Analysis: Machine learning generated analysis of the audio concerning the level of detected toxicity and disruption.

Listen to audio recordings#

Moderators can listen to audio recordings by selecting the incident number attached to a player report.

The audio is broken down by individual speakers. Each player has an audio track that is annotated with detected moments, which are detected interactions or situations that contribute to the risk scores. Moderators can mute individuals or listen to all the speakers at once to get the whole story behind a player report.

Understanding audio analysis results#

An overview of what information is contained within the audio analysis can be found in the following pages, starting with the Analysis.