Background

As we discussed in our previous post on user specific noise reduction, the personalized noise reduction software utilizes new methods to perform noise suppression. The results are impressive. The following recordings were done by executing the personalized noise reduction software under the toughest conditions possible: analyzing a single mono audio channel without any additional input like reference microphones, dedicated sensors, etc.

In order to view the below samples your browser needs to support HTML5.

[EDITED] These audio samples were created with a very initial version of this technology. You are invited to contact us to receive new samples that show significant improvement of this technology.

Mobile phone

Modern mobile phones are equipped with multiple microphones and dedicates sensors to help them attenuate the ambient noise. This built-in NR technology works well in many cases but fails in cases where the source of the noise is close to the microphone or is very loud. The following recording demonstrates how SoliCall’s personalized noise reduction software improves quality when executed after the audio was already filtered by the build-in NR of a mobile phone. We repeated these tests with multiple brands of modern mobile phones and got similar results.

Original

If the video does not show – use the direct link.

After Filter

If the video does not show – use the direct link.

Street Noise

Making high quality calls while you are located close to a noisy street is a big challenge.

Original

If the video does not show – use the direct link.

After Filter

If the video does not show – use the direct link.

Kids

Have you ever tried making business call from home while your kids are around ? If you did, then you will appreciate a noise reduction technology that can attenuate the sounds that are made by your kids.

Original

If the video does not show – use the direct link.

After Filter

If the video does not show – use the direct link.

Music

Music in the background ? The personalized noise reduction can distinguish between the music and the voice of the speaker and attenuate the music. In order to demonstrate that the personalized noise reduction technology is applicable to all genders and/or languages this recording was performed with a female talking in a non-English language.

Original

If the video does not show – use the direct link.

After Filter

If the video does not show – use the direct link.

Open Space

A popular scenario is the need to make calls from an open space or crowded office. In these cases the level of ambient noise can be very disturbing.

Original

If the video does not show – use the direct link.

After Filter

If the video does not show – use the direct link.

Implementations

Since the personalized noise reduction does not require any specific hardware, it can be executed on any communication device (e.g. the mobile phones) or on the network. Even if the communication device already has a built-in noise reduction, as we have seen in the mobile case above, the personalized noise reduction can still improve quality since it is using unique methods for recognizing and attenuating background noise.

The next post, uses a simple scenario to discuss the difference between personalized noise reduction software and all other solutions for noise reduction that exists in the market today.

[EDITED] This is an old post and by now our technology has significantly improved. Please contact us for latest results.