How Does SmartCompress Work in Sonic Captivate?

Question

Answer

SmartCompress uses two SNR level estimators, compared to one in traditional systems. In Figure 1, the green represents the fast phonemic estimator, and the blue represents the slow, long-term estimator. You can see that as noise is changing with speech, there are dramatic differences between green and blue. SmartCompress analyzes the signals simultaneously and provides an accurate measurement of environmental changes in real time.

Figure 1. Graphic representation of fast and slow/long-term estimators in SmartCompress.

Speech Variable Processing (SVP) with SmartCompress optimizes all other listening environments. SmartCompress allows the system to intelligently overcome challenges associated with a rapidly changing listening environments for a natural, instinctive hearing experience, eliminating the need for Environment Classification with SmartCompress. In the real world, our listening environment is ever changing. For example, we could be attending a meeting in a conference room with people around us having side conversations. We might be participating in a conference call on speaker phone and there may be static on the line. SVP with SmartCompress optimizes for all of these situations and delivers the solution on many different levels: 

• It's effective: SmartCompress specifically addresses the problem of applying compression in noise.
• It's adaptive: An adaptive compression system overcomes limitations of a fixed Environment Classification system
• It works well with others: SmartCompress complements SVP, directionality and noise reduction, and any fitting rationale of your choice.
• It's accurate: SmartCompress precisely measures sound changes in the environment. It is able to detect the short-term SNR of the signal at phonemic speed, as well as detect the ongoing long-term SNR of the overall environment. This management of both short- and long-term SNR is what separates Sonic from our competitors.
• It discriminates: SmartCompress knows what is speech and what is noise. 
• It's intelligent: SmartCompress applies different amounts of gain and compression on the signal target speech versus noise.
• It's fast: SmartCompress detects the SNR in real-time. It recognizes and addresses noise that occurs in between speech phonemes and speech pauses. 

SmartCompress controls compression in speech-in-noise environments, as well as adaptively varying the compression in speech-in-noise environments. It also determines how much the compression needs to be decreased, or made more linear, for less noise in the output. Furthermore, SmartCompress controls gain in non-speech environments. It adaptively limits gain when speech is not present, providing comfort to the patient. Sometimes, the noise floor or the overall amplification can be overwhelming. If your patient isn't comfortable with their hearing instrument, they're not going to wear it. SmartCompress controls gain in non-speech environments, adaptively limiting gain when speech is not present. It determines the occurrence of quiet and noise-only situations and applies less gain to the input to keep things comfortable. It's two systems working together.

This Ask the Expert is an excerpt from the interview, Captivate - What Life Sounds Like. For more information, please visit www.sonici.us or the Sonic Expo Page on AudiologyOnline.