Action Injury Law Group | "Reverse Determination of Legal Privilege from Historic Calls"
Mr. James Gibson spent over 29 years in maximum security prisons for a crime he did not commit, before an appellant court ruled that he had been tortured into confessing, and falsely convicted.
After Mr. Gibson was exonerated, he retained Andrew M. Stroth and Action Injury Law Group to advance his civil rights lawsuit against a number of officers and the City of Chicago.
When Action received a production of over a thousand of Mr. Gibson’s phone calls whilst incarcerated, in a compressed lossy format and with no dialled number metadata available, Action turned to Intelligent Voice Limited to enable, on a pro bono basis, a rapid reverse-determination of client-attorney privilege from spoken content.
Intelligent Voice Limited’s GPU-accelerated speech to text returned the entire dataset for review the next working day, with an auto-generated executive summary of each recording then enabling the review team to confirm privilege in a matter of seconds for any given call.
Alvarez & Marsal Disputes and Investigations | "Standalone Private Audio Transcription for Complex Matters"
Alvarez & Marsal Disputes and Investigations Germany implemented a standalone audio transcription solution with Intelligent Voice for a complex, privacy sensitive matter.
Intelligent Voice provides the same processing capability and functionality whether cloud or server based, and for air-gapped workflows between processing and review environments, as was the case here.
This enabled a large audio dataset to be rapidly processed multilingually and at scale for a combination of Germanic & Romance Languages.
Alvarez & Marsal managed to process and analyze a vast and intricate audio dataset, which significantly improved efficiency.
Consilio | "Continuous Active Learning for Enterprise Legal Solutions"
Intelligent Voice, a leading specialist in voice and analysis solution, provides end-to-end workflows for regulatory investigations, litigation support, Data Subject Access Requests, eDiscovery / eDisclosure, and combinations thereof.
Intelligent Voice’ capabilities include language detection and switching for multilingual matters, concurrent audio redaction sets for distinct disclosures of overlapping data, and full support for Continuous Active Learning, for both enterprise and managed service legal solutions.
Intelligent Voice’s high-volume audio and video capabilities integrate into proven workflows for maximizing the quality of the entire review process, with accelerated, scalable processing and daily utilisation reporting.
HaystackID | "Conceptual Analytics for Multi-Jurisdictional Investigations"
HaystackID, the trusted eDiscovery partner for corporations facing complex, data-intensive investigations and litigation, successfully deployed Intelligent Voice’s end-to-end audio discovery processes for one of the largest financial and banking institutions.
Even where a reviewer may have struggled to discern precisely what was being said, and so whether a particular call merited further investigation, techniques and technologies developed by Intelligent Voice enabled an accurate, quick and defensible human assessment of the dialogue.
Despite the age and quality of the audio recordings, the range of accents across a number of different territories, and the terminology sometimes used on the calls, Intelligent Voice’s industry-leading accuracy and retrieval enabled Relativity Conceptual Analytics to be successfully deployed, at scale, for a high volume, high value, multi-jurisdictional matter.
RelativityOne Justice for Change | "Unpicking Police Interview & Interrogation Recordings"
For attorneys working on a pro-bono basis to help the wrongfully convicted, having to dip in and out of recordings as they revert from their billable caseload and then back again can be challenging.
A navigable executive summary of interesting things said, provided as a sidebar in the review player, helps to mitigate, enabling an overall sense of the interaction to be quickly obtained and retained for a given pro-bono reviewer.
Interrogation recordings can present a particular challenge, due to the repetitive yet non-identical nature of questioning both within and across recordings for a detainee, some two or three hours at a time.
The executive summary provided by the automatically generated Intelligent Voice tags enables repetitive topics to be rapidly identified and sequentially reviewed.
The Intelligent Voice tags can also surface potential blind spots and unknown unknowns not mentioned in a case summary or report.
RSM Italy Forensic Investigations & Intelligence (formerly 36Brains) | Harnessing AI for Multilingual Audio Evidence Review
36Brains, a corporate intelligence company with offices in Milan and Berlin, offers intelligence and investigation services including forensic investigations, reputational assessments and litigation support.
The company’s CEO has led some of the most delicate and high-profile private intelligence operations of our time and is a pioneer of cyber investigations in the Italian market.
Central to the 36Brains ethos is the disruption of the corporate intelligence and investigation offering through the combination of old-school intelligence work and technological innovation.
In the context of a high-profile humanitarian criminal trial, 36Brains partnered with Intelligent Voice to rapidly identify and exclude tens of thousands of hours of inadmissibly poor quality covert in-vehicle and offshore recordings.
36Brains leveraged Intelligent Voice’s unique GPU speech recognition to visualize data across multiple languages simultaneously, ensuring effective review.
S2|Data | Enhanced Intelligibility of Legacy Video & Audio Recordings
S2|DATA provides legacy data solutions to some of the world’s leading organizations, enabling them to have total command of their legacy data whilst eliminating risk. Partnering with Intelligent Voice®, S2|DATA provided highly accurate transcriptions from decades old, borderline intelligible, audio & video tape recordings.
S2|DATA’s experienced data forensic experts managed the identification, acquisition, preservation, and digitisation of the legacy voice recordings, without compromising the data or risking its defensibility.
Intelligent Voice transcribed and converted the digitised video recordings into its platform-independent, NVIDIA award winning Intelligent Voice Review Player format. This provided both a navigable tag summary and a split-gap separated, karaoke-aligned and highly accurate transcript for occluded, far field, borderline intelligible speech captured on turn-of-the-century consumer audio tape recorders and video cameras.