
Intelligent Voice Redaction by Intelligent Voice
Used by regulatory authorities, law enforcement agencies, forensic eDiscovery specialists, law firms and in-house compliance monitoring and eDiscovery teams, Intelligent Voice is the only audio solution that covers the combined Relativity use cases of Redactions and Data Subject Access Requests.

Supported Audio Redaction Scenarios
Intelligent Voice supports multiple concurrent redacted versions of the same audio / video file within Relativity. As such Intelligent Voice in Relativity supports all combinations of the following 6 redaction scenarios:
- Teams with different expertise, e.g.: one team focuses on profanity or irrelevance, whilst another team focuses on assessments of Privilege
- Hierarchies within a team, e.g.: where paralegals apply draft redactions for approval, and reviewers confirm and approve
- Different requesting parties, e.g.: multiple parties on the same call recording, with the attorneys for different parties requesting different redactions to the same recording
- Sequential or overlapping requests from the same party, e.g.: a DSAR by the party alongside a pending case disclosure regarding the party or to the party
- Separate regimes & redaction rules within a given jurisdiction, e.g.:
- Disciplinary
- Regulatory
- Litigation
- Criminal Justice
- Variations in privacy requirements across different jurisdictions, e.g.:
- New York versus Frankfurt
- California versus Texas
Any and all of these 6 scenarios are concurrently supported by Intelligent Voice within Relativity.
As such there is no need wait until another user or team has produced their redacted media or/and text, or had the review of their own redactions completed, or even applied all of their redactions to any given recording.
Audio Redaction Functionality
Authorised users can create an audio redaction overlay by either:
- Clicking and dragging across the desired word or words in the automatically transcribed index – or
- Entering sub-second start & end timestamps
Users can then preview files with redactions applied, to ensure the file has been properly redacted, prior to production.
A redacted copy of the native file is not needed to enable a separate user to review and / or amend, or to save a user’s own changes between sessions.
Audio Redaction Workflows
A redacted copy of the native file is only required if redacted words are to hidden from dtSearch etc. or / and if a redacted version of the file is confirmed as being both fit for and required for production.
Each step in the redacted production is available as a Single File or Bulk Process, or as combinations of the two, independently of any other step in the process:
- Creating a Redacted Copy of the Native file
- Swapping the Redacted Copy in to become the Native
- Produce the Redacted Copy
- Revert the Native to the original Collected file
Redacted files integrate with Relativity’s own native production processes for both single file downloads and bulk production sets, whether for media, text, or both.
As such both redacted and unredacted objects can be combined in the same single Relativity production set, with the same production processes followed whether files have been redacted or not.
Audio Redaction Pricing
- The only cost from Intelligent Voice Limited is for the speech to text processing undertaken prior to redacting.
- If speech to text processing had already been undertaken for Analytics or Search, then redactions become available at no additional charge.
- If speech to text is required specifically to enable redacted productions of media and / or text, a to-the-cent pre-processing billing assessment enables the cost to be determined prior to any processing.
- Free trial of both speech to text and redactions available for any RelativityOne customers in the US.
- Free use for approved Relativity Justice for Change projects, leveraging the unique capabilities of Intelligent Voice and Relativity to positively impact racial and social justice in our communities.