Skip to content

Nuix Voice

Nuix Voice powered by Voci is an advanced machine-learning technology that converts electronically recorded speech to text. It integrates Voci’s V-Discovery speech analytics platform with Nuix’s portfolio of products for eDiscovery, investigation, information governance, cybersecurity, and intelligence. Nuix Voice turns recorded and live speech into accurate and fully-punctuated transcripts for analysis, search, and additional downstream processing.

With Nuix Voice powered by Voci, you can search and analyze all this human speech alongside emails, text messages, documents, chats, and many other sources.

How does Nuix Voice work?

Conduct Full-spectrum Communications Analysis

Combining Nuix Voice with the Nuix Engine enables you to perform text analytics and search across all communications—emails, text, chats, recorded phone conversations—and build accurate, content-rich timelines and communication pattern analyses.

Capture Every Detail

Nuix Voice can be configured to automatically detect specific keywords or phrases during processing. It also provides all manner of additional metadata including sentiment and emotion, gender of speaker, silence, time stamped words, confidence ratings, and speaker separation.

Convert Hours of Speech in Minutes

Nuix Voice can convert more than 100 hours of recorded audio or video to text in just one hour of clock time on a single appliance. That means you can work through hours of archived recordings and voicemails or monitor conversations in real time.

Deliver Audio Search for Everyone

Until now, audio search and transcription have been time-consuming and cost prohibitive. With Nuix Voice you can automatically convert the speech in every audio or video file to text for faster and easier analysis.

About the technology

Voci’s speech-to-text technology breaks speech out into words, language, dialect, emotion, and speaker. Voci uses a speaker-independent, large vocabulary continuous speech recognizer (LVCSR) to perform speech recognition. The LVCSR comprises two parts: an acoustic model that focuses on how things sound and a language model that focuses on words and word sequences. Voci’s most evolved language models are currently for North American English, UK English, and Australian English. It is also available for North American and European Spanish, with more languages on the way. Nuix Voice has closed the gap on speed against phonetic search without the false positives, false negatives and dictionary dependence.

Nuix Voice report

Deployment Options

You can deploy Nuix Voice as:

  • A cloud service for simplicity
  • An on-premise appliance for security and performance
  • An on-premise virtual machine for security and convenience.

Supported Formats

Nuix Voice supports more than 350 formats including:

  • Audio—mp3, NMF, G.7*, GSM 6.10, WMA, and many more
  • Video—AVI, mp4, wmv, and others as needed
Volver arriba