earSmart™ eS320 Advanced Voice Processor
The Audience earSmart eS320 is the first Advanced Voice Processor dedicated to the mobile computing segment, delivering a best in class voice experience for devices such as ultrabooks, tablets and all-in-one (AIO) PCs.
With the increased popularity of Voice over IP (VOIP) communications, as well as the use of voice as a means of interface, command and control, it has never been more important for these mobile computing platforms to employ robust, reliable voice processing technology.
360o Voice | Full range of movement around device with no loss of voice quality
Convertible and detachable Ultrabooks and AIOs represent a fundamental shift in PC form factor design. This design change poses significant acoustic challenges for the capture of clean, accurate speech since the speaker’s orientation to the unit can vary greatly. Legacy beamforming technology requires the speaker’s location to be fixed with respect to the microphones – usually it requires the user to be sat directly in front of the device - something that cannot be predicted when the device can be held or placed in a wide variety of positions.
With Audience’s 360º Voice, the earSmart eS320 uses unique ‘omni-directional far talk’ technology to deliver increased consistency of voice quality and increased reliability of Automatic Speech Recognition (ASR). Users benefit from a greater range of movement with respect to their mobile computing device.
Super Wideband Noise Suppression for VoIP | Make and take ultra-clear calls in nearly any environment
The earSmart eS320 features third generation Advanced Voice technology that supports Super Wideband VoIP calls (24 kHz) in addition to Narrowband (8 kHz) and Wideband (16 kHz) noise suppression.
The earSmart eS320 also features Acoustic Echo Cancelation that significantly reduces the echo effect that can occur during video calls, and the world’s first implementation of de-reverberation technology for mobile devices, significantly reducing the echo effect heard in cavernous spaces.
Optimized Speech Recognition for Ultrabooks & Tablets | Works reliably in noisy environments on range of platforms
ASR Assist technology directly addresses the challenge faced by many speech recognition applications today, namely their failure to recognize spoken words and complete assigned tasks because of disruptive background noise. The earSmart eS320 uses custom hardware-accelerated algorithms to isolate voice from surrounding environmental noise, which can improve the accuracy of speech-enabled applications such as voice search and speech to text by up to 40%.
Mobile computing devices equipped with Audience ASR Assist technology deliver improved speech application reliability, accuracy and task completion – even in noisy, distracting environments.
Acoustic Design Flexibility
Legacy solutions for noise suppression on mobile computing devices rely heavily on the user’s position in relation to the microphones. Audience’s CASA technology uses many other cues such as pitch, frequency, intensity, onset, spatial location, and duration to process voice information. By design, it can suppress noise coming from any direction.
This robustness not only allows for flexibility in microphone placement locations, but can also result in a reduction of material costs for the OEM, as use cases that require 3-4 microphones with beamforming, only require 2 microphones with the earSmart eS320.
The Science Behind the earSmart eS320 Advanced Voice Processor
- Non-Stationary and Stationary Noise suppression (Far talk)
- Automatic Speech Recognition (ASR) Assist mode
- 360 Voice - Omni-directional Far talk for increased positional robustness and Mic placement flexibility
- Audience Acoustic Echo Cancelation (AEC)
- Automatic Gain Control, Voice Equalization, Post Equalization and Multiband Compressor
- Narrowband (8kHz), Wideband (16kHz) and Super Wideband (24 kHz)signal processing
- Transmit De-reverberation
Key Device Features
- Third Generation Audience earSmart Advanced Voice Processing core
- 40nm process technology
- 5x5 mm footprint
- Digital Audio Ports
- 4 Master/Slave PCM/I2S ports
- Supports 4 digital Mic inputs and 1 output
- Up to 192kHz sample rates
- I2C and UART Host Interfaces