Nuix partnership adds speech transcription and voice analytics
Australian technology company Nuix and voice analytics provider Voci Technologies have announced a partnership in which Nuix will license and distribute Voci’s V-Discovery speech analytics platform. The two companies will integrate Voci’s V-Discovery engine with the Nuix Engine and Nuix’s products for ediscovery, investigation, information governance, cybersecurity, and intelligence.
“Organisations and individuals are recording massive quantities of audio and video and it’s impossible for investigators to listen to even a fraction of it,” said Nuix Chief Technology Officer Stephen Stewart.
“They’re accumulating audio for surveillance and compliance, voicemail, interviews, and on a huge range of devices including smartphones and police body cameras. Combining Nuix’s and Voci’s world-class technologies means customers can now search and analyse all this human speech alongside communication patterns, emails, text messages, documents, chats, and many other sources.”
Voci's V-Discovery system uses advanced machine-learning technology to convert electronically recorded audio data to text. A single Voci appliance can process more than 100 hours of audio in one hour of clock time, providing accurate and fully-punctuated transcripts of recorded or live audio sources in English or Spanish, and many more languages to come in the future.
The partnership allows organisations to:
- Perform a full-spectrum analysis that includes speech content as well as acoustic measurements of participant gender, emotion, and conversation times;
- Convert voice mail, recorded phone calls, interviews, video files, and other audio sources to text over 100 times faster than listening;
- Analyse human speech in real time or in batches while “listening” for specific phrases and automatically indexing audio files based on keyword occurrence; and
- Scale the use of this technology to any size organisation to provide audio search across the enterprise by delivering a cost-effective means of converting recorded human speech to text.