Fortemedia Advanced Microphone Array Processing(AMAP)

Fortemedia's Advanced Microphone Array Processing (AMAP) is a hybrid voice processing system that combines spatial filtering, source separation and adaptive statistical signal processing to deliver advanced voice and audio capabilities to consumer electronic devices.

How It Works?

As depicted in Figure 1 following, Fortemedia's AMAP utilizes a combination of spatial signal processing, signal separation and adaptive filtering to provide critical voice processing functions. AMAP supports single microphone and multiple microphone configurations. Spatial diversity of multiple microphones provides the possibility of separating a speech of interest from an interference which arrives via a different acoustic path, regardless the type of noise source - both constant and time varying. The underlying theory is that the advanced spatial filtering in the microphone array that identifies the locations of different sound sources and applies attenuation on the path of noise or interference. When combined with source signal separation algorithms and advanced statistical filtering techniques this spatial signal processing is the first step in a complex integrated process that is employed by AMAP to perform ambient noise suppression, acoustic echo cancellation and other voice processing functions.

Figure 1 - Fortemedia Advanced Microphone Array Processing System Diagram

The spatial filtering system integrates with a time-frequency processing system. The time-frequency processing system generates cues and parameters including a group of voice activity detectors that monitor the speech activities across frequency. Statistical variables for both noise and speech are also estimated in this stage. The time and frequency system incorporates voice pitch and sound location direction-of-arrival information, to separate target speech from background echo and interference. Voice pitch extracted from monaural input exhibits unique spectra structure and viable frequency range, interacting with direction of arrival information and other cues enabled by the multiple-microphone input, it provides the footprint to speech of interest. Such valuable information achieves a good pinpointing of far-end echo, ambient background noise and spot interference and helps their segregation from target speech.

In the final stage of the AMAP process, the spectrum of the main microphone input after spatial enhancement will be modified by gain factors across frequency to enhance the speech elements and suppress noise components. These spectrum modifications are performed sequentially on each concurrent output of the advanced statistical analysis filter and the corresponding processed frequency bins are reconstructed by the synthesis filter as the final output.

Upon those spectral signals, statistical filtering for noise suppression, and echo cancellation are conducted along with generating the voice activity statistics extracted previously. After stages of linear and non-linear processing, spectra signals are then converted back to temporal domain signal by a frequency-to-time mapping engine as clean speech output.

Attributes of AMAP

Scalability - from 1 to N microphones - AMAP technology is scalable from 1 to multiple microphone systems using the same core signal processing. This enables developers to employ AMAP in a variety of configurations using the same software base. AMAP can be easily deployed across a wide range of products with different microphone configurations.

Efficient computation for low power consumption - emphasis has been placed upon computation efficiency which results in the lowest power consumption possible for the features offered by AMAP. This computational efficiency allows AMAP to be deployed on a wide variety of platforms including CODECS with DSP, shared DSP and even ARM based architectures, all providing low power consumption profiles.

Flexible latency voice/audio processing - the AMAP process has flexible and programmable latency options allowing the technology to be employed in any communication network from LTE, 3G to VoIP networks. No special accommodation is required for AMAP to be used in these networks.

Fast convergence for noise suppression - resulting in excellent speech quality with minimal artifacts and noise leakage into the clean voice signal path.

Multiple spatial regions for voice isolation - enables multi-user applications with benefits of the voice processing for all users. Allows development of specialized audio and voice processing features such as dual user "interview" mode or "desktop speakerphone" mode with multiple users.

With AMAP Fortemedia provides critical communication enhancement functions for mobile devices functions consisting of:

Ambient Noise Suppression - AMAP provides strong noise suppression of ambient noise excluding sounds identified as out of the field of interest and by accurately tracking the voice signal of interest and eliminating other distracting sounds.

Far Field Voice Pickup - for conferencing applications of all types, AMAP provides improved pick up and isolation of remote voices. AMAP allows for effective voice pick up to distances of 3 meters and greater. Additionally, using the multiple microphones to establish a voice pick up zone that is remote from the device, AMAP provides improved speech recognition for hands-free consumer devices of all types - mobile phones, tablets, PCs, auto infotainment systems, wearable devices, gaming and TV systems.

Improved Acoustic Echo Cancellation (AEC) - Fast-converging linear filtering and smooth non-linear spectral modification constitute proprietary AMAP AEC/Noise Suppression technologies that result in excellent acoustic echo cancellation performance.