Voice Enhancement

Extended Voice Communication Package

About

Extended Voice Communication Package (eVCP) is a universal software package of digital signal processing technologies for voice applications enabling high quality, full duplex, noise- and echo-free communication in multi-microphone voice communication systems. eVCP is based on Alango Voice Communication Package (VCP) and extends its functionality by enabling multi-point voice communication in multi-zone, multi-directional, and/or multi-user environments.

eVCP supports distributed microphone configurations with single- or multi-microphone arrays at each voice communication point (zone).

Examples of such environments include:

• car cabin with multi-seat / multi-zone capabilities
• conferencing systems with multiple microphones located at different physical locations
• multi-directional desktop speakerphones and portable conferencing solutions

eVCP Components

eVCP is a complete voice enhancement solution that offers all required processing such as Acoustic Echo Cancellation, Noise Suppression, Beam-forming, Gains management and Intelligent Mixing. Acoustic path of each individual zone of a multi-zone voice communication system is processed separately to enhance its signal in the most optimal way, and the multitude of individual echo- and noise-free enhanced signals are combined into a signal enhanced uplink (TX-out) signal using the Intelligent Mixing (IM) block.

The Intelligent Mixing block is user- or manufacturer-adjustable and allows selecting a desired operation mode among the following:

• All zones active: for example, both the driver’s and passenger’s voices are preserved, while ambient noise and echo are attenuated.
• Selected zone(s) active: for example, the driver’s voice is preserved while the passenger’s voice, ambient noise and echo are attenuated.

eVCP includes the following signal processing blocks:

Two or more microphones in the device are positioned to adaptively discriminate between speech and noise utilizing Alango's beamforming algorithms. The algorithm supports multiple types of beam-forming including end-fire, close-talk, blind source localization and broad-side.

Eliminates acoustic echo with multi-band residual echo suppressor ensuring full-duplex communication with minimal voice distortions. Mono and Stereo AEC are supported.

Detects and attenuates stationary and transient noises (traffic, pub, passing cars, etc.) in transmitted and received signals.

Equalizes voice levels in the microphone and loudspeaker signals ensuring clear intelligibility. Alango AGC technology is robust to ambient noises and allows fast signal equalization without increasing ambient noises.

Allows fine frequency adjustment (equalization) and limiting of microphone and loudspeaker frequency responses with high frequency resolution.

Modifies the loudspeaker signal according to the ambient noise level and spectrum providing perceptually equal loudness and intelligibility in dynamically changing conditions.

• All zones are active
• Selected zones are active with other zones suppressed

Technical Information

PERFORMANCE

• Virtually unlimited number of voice zones
• Supports single- or multi-microphone per-voice-zone configurations, with omnidirectional or unidirectional mics, or multi-mic beamforming arrays (end-fire or broad-side beamforming)
• Echo canceller
  • Echo suppression with high frequency resolution (up to 31.5 Hz per band)
  • Fast convergence (< 300ms) with no initial echo
  • Convergence in double talk and high noise
  • Robustness to speaker signal distortions
  • Long echo tail (up to 1000ms)
  • Residual echo level < -60dB
• Noise suppression
  • Noise suppression with high frequency resolution (up to 31.5 Hz per band)
  • Up to 30dB noise reduction with low distortions
  • Fast noise adaptation (as low as 100ms for transient noises)
  • Single-channel and multi-microphone wind noise reduction
• Maximal AGC gain of 40dB
• Processing delay as low as 12ms
• Flexible processing frame size (including standard Bluetooth 7.5ms frame, 10ms for VoIP, etc.)
• Supported sampling rates: 8kHz, 16kHz, 24kHz, 32kHz, 48kHz
• Based on VCP which is fully compliant with with ITU/VDA/CarPlay and eCall/ERA-GLONASS automotive specifications

AVAILABILITY

eVCP is available* on the following platforms:

• ARM cores (all types including ARM M4 and M4F)
• CEVA TeakLite III, IV
• Renesas SuperH family
• Synopsys ARC cores
• Cadence (Tensilica) HiFi2, HiFi3, HiFi4, HiFi5
• Analog Devices SHARC

* Implementation and availability of specific eVCP blocks may differ between platforms. Please, contact Alango technical support for specific information.

Porting on other platforms can be performed quickly.

Intelligent Mixer Audio Demonstration

An audio demonstration of Alango's Intelligent Mixer technology can be accessed at the link below:

Download Intelligent Mixer Demo
(PPTX, 40.7 MB)

Please contact us for more information, current MIPS/memory consumption, audio examples and real time demo availabilities.