Voice Inspector

Voice Inspector Introduction

Phonexia Voice Inspector (VIN) is a powerful tool designed by Phonexia for forensic experts and law enforcement agencies to assist in criminal investigations through highly accurate speaker identification and. It performs automatic forensic voice comparisons in compliance with the ENFSI guidelines and helps prepare comprehensive case reports for court proceedings.

 

Top Three Benefits of Using Voice Inspector

• Reduced Investigation Time – Analyze the subject’s voice automatically with an advanced speaker identification tool, and support your forensic expert’s conclusion with accurate, unbiased voice analysis..
• Efficient Voice Analysis – Identify a speaker in the recordings of any language without the need to hire a language-specific linguist as Phonexia Voice Inspector can detect pronunciation differencies in any language.
• Detailed Reports – Present the results of your forensic voice analysis to a court in the most convenient way with an automatically generated report containing all the necessary details to validate the claim.

 

Ultimate Voice Analysis Software for Forensic Experts

 

Phonexia Voice Inspector is an out-of-the-box solution that provides police forces and forensic experts with a highly accurate voice comparison tool to support effective criminal investigations and give evidence in court. It leverages Phonexia’s cutting-edge voice biometrics technology for speaker identification, providing exceptional accuracy.

 

Powered by State-of-the-Art Deep Neural Networks

 

By leveraging deep neural networks, the latest generation of Phonexia Voice Biometrics called Deep Embeddings™ is the most accurate speaker identification technology available on the market. Phonexia Voice Inspector utilizes this technology and provides you with an intuitive graphical interface so that you can perform advanced forensic voice analyses with ease.

 

Key features:

• Adaptive Voice Comparison: Whether you need to verify a pair of voice recordings against each other (1:1 identification) or search for a speaker within multiple audio files (1:N identification), Phonexia Voice Inspector allows you to do both so that you can choose the right approach necessary for your case.
• Language Independent: Compare voice recordings regardless of their language and eliminate the need to hire a dedicated linguist specialized in a particular language as Phonexia Voice Inspector can identify the speaker’s unique voiceprint in any language, making forensic analysis more efficient.
• Automated Unbiased Analysis: Reinforce your forensic claim with an objective voice analysis done in Phonexia Voice Inspector natively using Phonexia Deep Embeddings™—the latest generation of automatic speaker identification technology powered by deep neural networks to provide high accuracy.
• Multiple File Search: Analyze large amounts of audio recordings with ease using Phonexia Voice Inspector’s built-in ability to search for identical phoneme sequences across multiple voice recordings so that you can work more effectively and provide forensic voice analysis on time.
• Wave Editor: Cut through the noise quickly with Phonexia Voice Inspector’s Wave Editor, which lets you automatically detect the audio parts containing speech, flag the recordings unsuitable for voice analysis due to their noise level, display a spectrogram for more detailed analysis, and much more.
• Easy Case Management: Stay on top of every forensic case with straightforward management of audio files, population sets, and corresponding notes so that you can progress through each investigation systematically and with confidence that all case-relevant files are always in one place ready for analysis.

 

 

Input requirements

 

• Audio: WAV (8Khz+ sampling)
• Population set: 10+ speakers (50+ recommended)
• Recordings: Suspected speaker (20 seconds of speech), questionned speker (3+ seconds of speech)

 

Output

• Scoring table: with comparison results via a likelihood ratio, log-likelihood ratio, and verbal ratio.
• Graphical presentation: in the form of a Proability Density Function plot and Tippett plot
• Detailed Customisable report: generated automatically and exportable as a PDF or open document file