Bongiovi Medical has conducted extensive research into improving speech intelligibility in communication systems, telemedicine systems and medical devices. To understand how we do this, one first must understand how we hear and perceive sounds.
It is generally accepted that the average human can hear sounds from 20 Hz to 20,000 Hz. This is know as the audible frequency range. Human beings, however, do not perceive loudness equally across the audible frequency range. Human hearing is “tuned” to be the most sensitive around 3,000 Hz. Sounds in and around this range have been key for survival – sounds of a predator stalking or a baby whimpering. With human speech, this is a very important range. However, there are higher frequency ques that also aid in understanding speech. Sounds outside of the 3,000 Hz range need to be much louder (higher sound pressure level, or decibels) to be perceived at the same loudness as 3,000 Hz. The graph below illustrates this very clearly.
As you can see, it takes considerably more power for 100Hz to sound as loud as 3000Hz.
Bongiovi Engineers’ understanding of how the brain perceives sounds enabled them to build unique “intelligibility” profiles for MDPS™. We can make human speech more intelligible. We can also make low frequency biological sounds easier to hear.
When it comes to the human voice, we make it clearer and sound louder. MDPS™ is able to increase intelligibility without increasing the actual sound pressure level in the playback environment. The Perceived Volume is increased, but the Actual Volume remains the same. To do this, we alter both the frequency content and dynamic range in real time. This maximizes intelligibility and audibility based on the way human beings actually hear and our brains actually interpret sound.
BMHT conducted an independent pilot study on a major university medical center’s telemedicine system. We intentionally created a less than ideal listening environment that replicates real world usage of the telemedicine system. Intelligibility was measured using the ANSI Modified Rhyme Test ANSI/ASA S3.2-2009, “Method for Measuring the Intelligibility of Speech Over Communication Systems”.
Below are the results of the pilot study: