Amazon Transcribe Toxicity Detection

Detect toxic content in spoken conversations

Why Amazon Transcribe Toxicity Detection?

Amazon Transcribe Toxicity Detection, an ML-powered, voice-based toxicity detection capability, which leverages both audio and text-based cues to identify and classify toxic language including hate speech, harassment and threats. In addition to text, Amazon Transcribe Toxicity Detection uses speech cues, such as tone and pitch to hone in on toxic intent in speech. Toxic content is flagged and classified across seven categories including sexual harassment, hate speech, threat, abuse, profanity, insult, and graphic. This allows moderators to take focused action rather than reviewing entire conversations. 


Review a select few or hours of audio conversations for offensive and innappropriate speech. Proactively provide a safe online environment for both users and brands. 

Automatically analyze audio conversations at scale to highlight areas with toxic content. Reduces the content, human moderators have to review by up to 95%.

Configure Amazon Transcribe Moderation by enabling a few options in the AWS console, no machine learning experience is required. 


Amazon Transcribe Toxicity Detection classifies toxic audio content and provides a confidence score (0 to 1) for the following seven categories: sexual harassment, hate speech, violence/threat, abuse, profanity, insult, and graphic. 

Easily filter results by managing threshold values for each toxic content category. 

Transcribe Toxicity Detection can analyze the following media formats: MP3, MP4, WAV, FLAC, AMR, OGG, or WebM formats.