Solution for Intelligence

Rapid detection of suspicious persons and criminal intent to ensure the security of the state and prevent crimes

Problem

Systems of technical data collection produce large data sets, among which it is necessary to quickly identify data indicating criminal intentions against the state and planning for their implementation, as well as to identify suspicious persons. 

The main problem is that data is collected very quickly and in large volumes and becomes outdated even faster. Detected suspicious scenarios require a quick response, so data processing time is very important.

The main purpose of using Harvester: quick filtering of the existing array of data in order to identify phrases and persons that may indicate a threat to the security of the state. All detected suspicious events must be transferred to the operator for their quick analysis and making appropriate decisions.

Solution description 

1

All the collected audio data goes through the stage of speaker Gender Identification and Speaker Recognition

Only those files that contain either a speaker of the desired gender or a speaker that is in the database of suspected persons are processed further. 

2

 

The necessary audio files are transcribed

3

The received text files are searched for critical words specified by the operator

The more different combinations of such words a file contains, the higher its rank.

4

The operator receives a database of audio and text files that correspond to critical scenarios

Next, the operator is able to evaluate the results and make a decision, if there are indeed evidence of crimes on the audio.

5

Machine Translation of the transcribed text into the required language

If the audio files contain a foreign language, it is also possible to make Machine Translation of the transcribed text into the required language.

6

Additional processing of files

Our expert designers work with you to get the right picture of your new website and to consider all the wishes and requirements. This process includes diverse approvals, tests, reviews, and close communication. 

Key Technologies 

Data types

Video & Images

● recognition of a person on the face (identification/verification) and analysis of portrait characteristics● people search● objects search and detection

Audio

● recognition of a person by voice (identification/verification)● speech transcription (audio to text conversion)● retraining of language models for transcription● speaker language identification in audio● determining the gender of a person in audio

Text

● language identification● translation of text documents● keyword spotting● sentiment analysis● text document summarization● named entity recognition● text documents● categorization by predefined topics

Try it out with your own data

Need few more text

Thank you!

We will contact you shortly

Can't send form.

Please try again later.

Made with