A new hearing instrument introduced by Sonic Innovations, called Innova, incorporates several new digital features derived from the principles of audio theory, the cochlear amplifier model, and digital signal processing for hearing aids. The new hearing instrument offers S.M.A.R.T. Technology, improved Noise Limiting algorithms, an advanced Feedback Cancellation algorithm, and a unique D3 Directionality function (Figure 1). The result of these features is what we believe is unprecedented benefit for improving speech understanding in noise, including AI-DI values that exceed 10 dB.
Figure 1. Functional block diagram of new Innova hearing aid, including features of dual-microphone fixed-configuration directionality, DIRECTIONALfocus algorithm, Feedback Canceller algorithm, 16-channel DSP as part of SMART Technology, and Adaptive Noise Reduction algorithm.
Monitoring the Listening Environment
Innova incorporates a suite of algorithms that constantly measures the users listening environment. When a change in the listening environment is detected, S.M.A.R.T. Technology analyses the new environment, then reacts within milliseconds to provide seamless transitions. Regardless of the environment, the hearing aid user no longer has to wait for their hearing aid to catch up with the listening situation; the instrument transparently adapts before important communication information is lost.
The new technologys patented design includes narrowband, wide dynamic range compression (WDRC) operating with very fast-acting time constantsthe foundation on which all other algorithms in the hearing system are layered. The new technology forms the basis for Innovas features, natural sound quality, and user benefits.
S.M.A.R.T. Technology provides 16 compression channels adjustable via 9 fitting handles. The handles are mapped to the 16 compression channels through the Accufit interpolation algorithm in the NOAH-compatible EXPRESSfit 5.0 programming software.
The compression time constants are different from those in other competitive hearing aids in three key ways:
The time constants vary by frequency;
The time constants are very fast, especially in the high frequencies, and
The attack and release times are symmetric.
The configuration of WDRC channels and distinctive time constants play crucial roles in the technologys ability to enhance speech cues while preserving the naturalness of sound.
In addition, non-traditional input/output (I/O) functions are designed to emphasize speech and preserve listening comfort for each individual patient. The I/O functions integrate low, middle, and high kneepoints in each compression channel, creating multiple line segments, each with unique slopes. These line segment slopes signify the rate of change in gain as the input level varies, and can be less than 1.0 (compressive), equal to 1.0 (linear), or greater than 1.0 (expansive). Thus, there are actually four line segments:
Expansion Segment: Reduces the gain for low intensity background signals;
Linear Segment: Provides maximum gain for important soft speech cues while limiting the feedback associated with high gain applied to low intensity signals;
WDRC Segment: Compensates for the reduced dynamic range of hearing associated with a sensorineural hearing loss; and the
Limiting Segment: Provides maximum output limiting with low distortion.
Gain Dependent Expansion with Adaptive Noise Reduction
The Noise Limiting system of the new hearing instrument combines the benefits of Gain Dependent Expansion and Adaptive Noise Reduction to combat both low- and high-level noise sources. The result is increased signal-to-noise ratio (SNR) performance across a wide variety of everyday listening situations.
Figure 2. Innova hearing instruments.
The new hearing instruments Gain Dependent Expansion addresses low-level noise, such as internal component noise or relatively quiet background noise, while maintaining amplification of important soft speech cues. The amount of attenuation applied to low-level noise varies based on each patients gain setting at each frequency, with more expansion applied to lower-gain settings and less expansion applied to higher-gain settings. Gain Dependent Expansion is designed to reduce bothersome noise at frequencies where hearing is still good, while preserving audible cues at frequencies where the hearing loss may be more severe.
Based on Sonic Innovations clinically-proven digital noise reduction algorithm, Innovas patented Adaptive Noise Reduction actively seeks out and reduces steady-state background noise based on the signal-to-noise ratio while providing more amplification for modulated speech targets. This new algorithm also adds the ability to regulate the amount of noise reduction based on the input level of the noise. As the input level of the noise increases, the attenuation of the noise becomes more aggressive. Adaptive Noise Reduction therefore tackles both low- and high-level steady-state noise sources, enhancing speech understanding in noise and providing more natural sound in many listening situations.
Feedback Cancellation Through the Use of a Counter-Phase Signal
The systems Feedback Canceller allows full use of the hearing aids fitting range without feedback, promotes the use of larger vents which reduces the occlusion effect, and makes it possible to address physical fit issues by allowing looser earmolds or shells that help alleviate physical discomfort.
Figure 3. Polar plot of the new BTE-D hearing aid. DIRECTIONALfocus mode is single speaker and fixed configuration mode is dipole. Articulation index weighted directivity index (AI-DI) measurement shows a value of 11.1, an exceptionally high value for dual-microphone, headworn hearing aids.
The feedback cancellation system uses a counter-phase signal to remove feedback before it becomes audible. The result is increased listener comfort without a reduction in fitting gain. During the fitting session, a training signal is played through the hearing aid, and the feedback path of the signal is measured. The hearing aid then calculates and stores values that are opposite in phase to the original signal. When the hearing aid is in use, the previously calculated counter-phase values are added to the input, thereby canceling the potential feedback signal.
The inclusion of the Feedback Canceller in Innova brings even more clinical programming power to the Voice Manager tool previously released in EXPRESSfit. The Feedback Canceller is designed to enhance the power of the Volume Manager, the Voice Position Manager, the Occlusion Manager, and the Sound Quality Manager to provide amplification without introducing uncomfortable changes in the quality of the patients own voice that so often happens during conventional hearing aid fittings.
Different Levels of Directionality for Enhanced AI-DI Scores
D3 Directionality combines several of the new hearing instruments key technologies to create a robust directional system with exceptional performance on AI-DI measures. The multiple levels of directionality are: D1 with configurable directionality, D2 with Adaptive Noise Reduction plus configurable directionality, and D3 with DIRECTIONALfocus plus Adaptive Noise Reduction and configurable directionality.
D1 Directionality is configurable among three available polar patterns (dipole, hypercardioid, or supercardioid) to provide benefit for a variety of listening situations. It is designed for higher noise situations, and emphasizes sound from the front while attenuating sounds from the sides or from behind the listener. The switchable feature allows the user to engage and disengage directionality as desired, depending on the informational value of the sounds from the sides and behind. Innova hearing aids are configurable with up to four independent directional programs. Depending on the listening environment, one polar pattern may be more suitable than another.
D2 Directionality is the combination of configurable directionality and Adaptive Noise Reduction. Designed for use in a variety of listening environments, this system is ideal for high, medium, and low noise situations. It emphasizes speech or music from the front and de-emphasizes unwanted steady-state background noise from any direction.
D3 Directionality combines the new DIRECTIONALfocus algorithm with Adaptive Noise Reduction and configurable directionality. It is designed to be the most advanced directional technology on the market, providing higher levels of performance in traditional measures of directionality. The DIRECTIONALfocus algorithm offers directionality through focused, adaptive gain manipulation based on the direction of soundwithout the need for the traditional time delay between the front and back microphones. Sounds in front of the listener are considered to be important; sounds originating from the sides and back of the listener are considered to be less important and are therefore attenuated.
D3 is, in a sense, a very focused form of directionality that is designed to be adaptive to any listening environment. It is similar to conventional adaptive directionality, in that sound originating from in front of the listener is fully audible; as the sound source moves to the side and behind the listener, the amplified level of the sound is automatically reduced. However, we believe it to be superior to conventional adaptive directionality because of the greater directivity index values. With DIRECTIONALfocus in D3 Directionality, AI-DI scores of greater than 10 dB are now possible, something previously thought impossible for head-worn, dual-microphone hearing instruments.
With the introduction of the original Natura hearing aid, Sonic Innovations brought sophisticated DSP to the market with 9-channel compression at a time when the competitive state-of-the-art was 2- and 3-channel analog and digital processors. Likewise, Natura 2 delivered digital noise reduction that is clinically proven to improve speech recognition in noise, at a time when the competitive state-of-the-art was noise reduction that only provides listening comfort. With the introduction of Innova, Sonic Innovations is bringing to the market an aid designed to enhance directional performance, with double-digit AI-DI values, in a time when competitive state-of-the-art is limited to single digits.
|This article was submitted to HR by Victor Bray, PhD, vice president for quality systems at Sonic Innovations, Salt Lake City. Correspondence can be addressed to HR or Victor Bray, PhD, Sonic Innovations, 2795 E. Cottonwood Prkwy, Suite 660, Salt Lake City, UT 84121; email: [email protected].|