Skip to main content

Speaker Diarization in VIDIZMO

VIDIZMO offers tools and features for conducting a detailed analysis of your media or evidence in your Portal. When transcribing audio and video files, the VIDIZMO Speech & Text Analyzer identifies and separates different speakers based on their voices through speaker diarization.

VIDIZMO accomplishes this by analyzing the unique characteristics of the speakers' voices, segmenting them, and assigning their transcriptions a speaker prefix (e.g., Speaker 1: ). The transcriptions generated reflect the identified speakers and their words, allowing you to quickly determine which sentences each speaker spoke.

The VIDIZMO Speech & Text Analyzer for diarization only analyzes the voice segments in the audio, making the feature language independent. This means the application can identify different speakers regardless of the language used in your audio or video files.

Prerequisites​

  • Ensure you belong to a group that has Transcription & Diarization and App Management feature enabled, or have a CAL that grants you these permission. By default, these feature are Portal-level add-ons that must be enabled in security groups.
  • The VIDIZMO Speech & Text Analyzer App needs to be configured for the Transcriptions AI Insight. For the configuration steps, refer to: Configuring VIDIZMO Speech & Text Analyzer for Transcriptions and Translations

Speaker Diarization Steps​

  1. Navigate to the audio or video you want to process.

  1. Click process on its overflow menu.

  1. Select 'Generate AI Insights' on the 'Process' window.
  2. Add 'Transcriptions/CC' to the Insights tab.

Note: Speaker diarization is performed whenever transcriptions are generated via this AI Insight. You can also generate transcriptions for your content in several other ways, refer to: How to Generate Transcriptions and Translations using VIDIZMO Speech & Text Analyzer

  1. Click 'Start' to begin the processing.

  1. Once the processing is complete, click your audio or video file to open its playback page.

  1. The transcriptions segmented by speakers for your file are present in the Transcription tab.

To learn more about transcriptions using VIDIZMO Speech & Text Analyzer, refer to Understanding Transcriptions and Translations Generation via VIDIZMO Speech & Text Analyzer.