IKAR Lab: Audio/speech signal analysis

Professional hardware and software set for advanced audio/speech signal analysis.

Since it has been launched in1992, IKAR Lab has evolved from an sound editor application to the most popular forensic lab in the world. Today it is serving experts in 350 laboratories in more than 60 countries worldwide.

Audio/speech signal analysis IKAR Lab
Product Library:

Overview

IKAR Lab is an excellent solution for speech signal analysis. Being the most popular audio forensic lab in the world it gives you powerful speech enhancement and audio restoration opportunities, sophisticated speech analysis capabilities, precise noise cancellation options and numerous editing tools. IKAR Lab was designed specifically for the extraction and in-depth analysis of voice parameters and is now installed in the most advanced forensic labs all over the world.

Our goal was to create a product with an optimal potential for audio processing that could meet all possible forensic needs and stand any intricate challenge.

Why choose IKAR:

  • powerful environment for noise cancellation and speech signal enhancement
  • transcription of low quality recordings
  • speaker identification
  • biometric voice database management system
  • authenticity analysis of analog or digital recordings

Features

  1. Match and catch

    Create, monitor and manage your voice database. You can now utilize all the recordings made during suspect interviews and investigative reports to create database of known or suspect voices in a passive manner and quickly identify or eliminate speakers under investigation.

  2. Analyze this

    Forensic analysis of audio recordings is now faster and easier. The combination of powerful tools for enhanced speech processing, audio visualization, segmentation, noise reduction and text transcription is now at your disposal as a unique software kit of great potential and value. Plus, you get two additional plug-ins that accomplish authenticity and identification analysis that all in all gives you an opportunity to provide clear files and text transcriptions at trials.

  3. What's that noise?

    Be it an accident, criminal investigation or 911call it is important for speech to be clear and easily understandable to ensure no vital information is lost. Solving the toughest tasks of speech enhancement and noise reduction, including real-time processing , we combined all up-to-date noise filtering algorithms in one product . Ikar Lab is also equipped with full-fledged noise-processing server that can be controlled remotely via web page in multiuser mode.

  4. Recording speakers

    Whenever there's need to take voice samples of speakers under investigation indoors or in the field our software and hardware solution is perfect for both. Stationary double microphone device with software interfaced to your database and a pocket shockproof voice recorder will embrace all needs and tasks to create your own voice sample database.

  5. Support

    Audio forensic training is a part of audio lab efficiency work. STC courses are designed to provide the student with detailed theoretical and hands on training on the suite of audio forensic products and methods used for different types of examination.

The structure of IKAR Lab

VoiceGrid Local

The VoiceGrid Local system provides the unparalleled voice biometric matching and voice database management capabilities for audio forensic lab. In 2010 Speech Technology Center successfully deployed the world's first nationwide voice-based identification system for the government of Mexico. It is the largest voice biometric identification project to date that has laid the foundation for a new segment in the voice biometric market with tremendous value and potential.

Forensic Audio Workstation

FAW is a complete software and hardware solution, which ensures comprehensive forensic analysis of analog and digital recordings. It consists of SIS II software with two additional plug-ins for authenticity and identification analysis and Sound Cleaner program that accomplishes filtering and noise reduction.

ANF II

ANF II is a noise filtering server while connected to the local network, the ANF II works as a full-fledged noise-processing server combining 9 filters. The device can be controlled and all the settings can be adjusted remotely via web page.

Voicesampling workstation

VSW is the complete software and hardware solution for taking voice samples of speakers under investigations that consists of:

  • STC-H216M with 2 mics

    Input/output audio device

    The device allows a audio output into the PC memory through the double microphone inputs and enables to listen to the audio information through the headphones.

    Specifications:

    • Connection interfaces: RCA analog line inputs/outputs
    • ADC/DAC resolution: 16 bits
    • Signal/Noise ratio: 82 dB
    • THD: 0,01%
    • Frequency response flatness: +/- 0,1 dB
    • Analog signal sampling rate^ 8-48 kHz
  • VoicePassport software

    • Automatically estimates voice sample quality
    • Automatically estimates “pure” speech duration
    • Transfers voice sample to VoiceGridLocal
    • Audio quality geared for successful voice ID
  • Gnome P Digital recorder

    Pocket stereo digital voice recorder

    Gnome P is used for taking voice samples in the field. The quality of audio recordings allows using them for audio forensic investigation. The recorder has the optimal size-function-quality ratio. Its metal casing provides powerful defense from mechanic deformations and electromagnetic emissions.

    Basic specifications

    • Built-in memory: 8 GB
    • Maximum recording time: 36 hrs
    • Built in Li-Ion battery

Support

Training: Forensic acoustics. Basic course

Duration: 5 days

  • Introduction into forensic audio
  • The theory of noise cancellation
  • Sound Cleaner: audio filters and their parameters. Practical application.
  • The theory of speech production
  • SIS II software interface, equipment testing, sound editing, processing and analysis
  • Forensic voice identification: theory and practice
  • Authentication
  • Making report

Automatic Voice Identification

Marking the Audio

Working with Project

Converting Spectrum into Filter