The Virtual Listener Panel™ (VLP) is your cloud-based solution powered by advanced AI, trained on extensive datasets collected through SenseLab’s expert listener panels.
Perceptual characterisation of your audio product 24/7
Specifically designed for audio professionals, VLP leverages SenseLab's trusted Sound Wheel attributes, enabling precise, attribute-specific characterization of your audio products around the clock.
Streamlined, data-driven audio optimization
With VLP, optimizing your audio attributes has never been easier or more insightful. Simply upload your recordings or simulations to instantly receive detailed perceptual analyses of key audio attributes. Explore how different tuning decisions influence specific sound characteristics, quickly and clearly seeing how close your current tuning is to a reference or benchmark product.
Why Choose VLP?
Objective, reproducible perceptual evaluations based on expert listener model
Fast, reliable attribute insights - scalable to your project's complexity
Immediate benchmarking against previous tunings or competitive standards
Perfecting your product’s audio performance
Take your audio product's perceptual characterization to the next level - efficiently and effortlessly.
Ready to experience VLP?
Request today and receive 300 credits to explore the future of perceptual sound characterization.
Contact SenseLab
Do you have a question, or want to discuss a project?
The VLP service is offered at three different subscription levels, starting with the Light version at 495 EUR per year (including 100 credits/month) up to Enterprise at 3,495 EUR per year, including 1000 credits/month.
The subscription covers access to the VLP platform, including the latest version of the model. The Premium subscription always allows you access to the newest model.
Every new analysis in VLP will be charged a certain amount of credits to run. You are informed how many credits an analysis requires before you accept or decline it to run. Your personal account has a counter on how many credits you have left. You can purchase more credits at any time.
The Virtual Listener Panel (VLP) is a machine learning-powered tool developed by SenseLab to evaluate and quantify the perceptual quality of audio products. It enables manufacturers to compare sound characteristics across different systems in minutes, as opposed to days or weeks when using traditional human expert panels. For more details click here.
The VLP can predict perceptual attributes in the timbre domain, including: Bass strength, Bass depth, Midrange strength, Treble strength, Brilliance, Timbre balance, Boxy and Nasal.
The VLP is continuously being updated to include new attributes for more predictions.
VLP can analyse perceptual differences based on attributes in the timbre domain. The model is trained to predict perceptual differences in music or noise playback.
The VLP uses a Siamese-inspired neural network model based on no-reference ratings, ensuring precise evaluations without the need for direct comparison to a reference file.
The VLP is trained on large data sets of paired comparisons collected with SenseLab's trained expert listener panel. Variations of sound samples were made specifically for each attribute and convolved with a range of real measured headphones and earbuds.
The VLP allows manufacturers to make perceptually informed decisions faster and more efficiently. By automating perceptual evaluation, it reduces the time and costs typically involved in testing with human expert panels.
The VLP uses a large dataset collected from SenseLab's expert panel, which includes around 54,000 ratings on music genre samples and pink noise. This data is transformed into 150,000 pairwise ratings for model training and validation.
The VLP model is trained using a Siamese-inspired neural network architecture. It has been validated to ensure its ratings align closely with real systems assessed by human experts.
Yes, the VLP is designed to be a scalable and easily integrable framework, ensuring that perceptually informed decisions are made from the very beginning of product development.
The VLP leverages a machine-learning approach based on no-reference ratings. It also employs extensive audio features, such as spectrum type and loudness normalisation, to enhance its testing precision.
