ADVANCED NOISE REDUCTION FOR MOBILE TELEPHONY
Â¢ Have you ever used your mobile phone to place a call from a noisy place and been asked to call back from someplace quieter
Â¢ It would be great if the phone would mute the background noise for you
Â¢ This can be done by advanced multi microphone noise suppression algorithm working with principle of human hearing system.
EFFECT OF NOISE
If the environment is noisy
It will enter into microphone so that it become difficult for the person on the other end of call to understand.
Voice encoder transmit unnecessary voice, so wastage of energy.
TYPES OF NOISE
Basically three types:
1. Stationary noise: Does not contain rapid changes eg: steady sound of a fan.
2. Non stationary noise: rapid changes
eg: music containing drums.
3. Quasi stationary noise: stationary for a particular time eg: steady babble of a crowd
PRESENT NOISE REDUCTION TECHNIQUE
Fast Fourier Transform:
Â¢ Removes noise which is stationary or quasi stationary.
Â¢ But fails to remove non stationary noise, because these noise changes too fast and are similar to the desired noise.
Â¢ Fast Cochlea Transform
ADVANCED NOISE REDUCTION TECHNIQUE
Based on the principle of human auditory system.
Processor used for this: Audience voice processor
Uses to two microphones to capture the information about the position of different sound sources.
So that phone can lock the desired voice and remove all others.
FCT vs FFT
1. FFT transforms the audio signal into the frequency domain on a linear scale, while the FCT performs its transformation on a logarithmic frequency scale.
2. The FCT provides greater accuracy in representing the audio signal at both low and high frequencies.
3. The FFT transforms the audio signal by reading blocks or frames of data that are taken at a particular frame rate. The FCT continuously streams the incoming signal into the transformation.
Processes in voice processor
1. Fast Cochlea Transform
5. Inverse Fast Cochlea Transform
FAST COCHLEA TRANSFORM
Convert sounds into a three dimensional, high-quality spectral representation of the sound mixture
hints for the different sound components are computed which include:
1. Pitch: a male voice and a female voice can be easily separated using Pitch
2. Spatial location: to locate the direction from which a sound is coming and its distance to each of the microphones.
3. Onset time: Frequency components from a single sound source often start and/or stop at the same time.
GROUPING & SELECTING
A grouping is process of clustering operation in which signals with similar attributes are grouped into one auditory stream and signals with dissimilar attributes are grouped into another auditory stream.
The Selector process allows the separated auditory sound sources to be prioritized and selected as appropriate for the given application.
INVERSE FAST COCHLEA TRANSFORM
The Inverse Fast Cochlea Transform process converts the Fast Cochlea Transform data back into reconstructed, cleaned-up, high-quality digital audio which is then converted back to an analog signal, and made available for transmission.
1. Instantaneous noise suppression.
2. Advanced acoustic echo cancellation.
3. Ease of integration.
4. Low power consumption.
5. Extensible architecture.
Thus the introduction of Audience voice processor will be a revolution in the field of mobile.
Thus we can have a clear and noise free calls