Harry Levitt, PhD, is CEO and Chief Scientist at Sense Synergy, where Chris Oden, is President/Chief Technology Officer, and Carla Noack, is a Research Assistant. Helen Simon, PhD, and Al Lotze are members of the Smith-Kettlewell Eye Research Institute in San Francisco, CA. Read My Quips was evaluated by Simon and Lotze under a subcontract awarded to the Smith-Kettlewell Eye Research Institute by Advanced Hearing Concepts.
Auditory training takes commitment, not just from audiologists who have to work intensively with patients over long periods of time, but also from patients themselves who have to spend hours improving their listening skills. The problem? Many auditory training programs do not live up to this standard, and patients often do not complete the program.
Now, though, new computer-based programs are overcoming many of the barriers that have prevented the use of auditory training, reducing dropout rates and improving the effectiveness of auditory training. The key has been to make the process more engaging by using computer programs that provide face-to-face communication in noise while helping users improve speech comprehension skills.
Read My Quips, for example, uses a format similar to a crossword puzzle, except each box in the puzzle contains a word, not a letter, and the words form witty quips. (Figure 1.) The clue for each item is a video recording of a person saying the quip against background noise, with the level of background noise adjusted automatically to match each trainee’s ability to understand speech.
The program can be accessed from any computer and even an iPad with Wi-Fi. A few of the boxes contain text, representing items that have been solved. To solve another quip, the trainee clicks on the row or column of the puzzle containing the item, and a video clip of a person saying the quip begins to play (www.sensesynergy.com/readmyquips). The first few words show in text below the speaker's face, and the trainee's task is to identify the remaining words of the quip that are shown as empty boxes. If a wrong word is entered or the trainee cannot recognize the quip, the video clip is played again with the background noise reduced. The trainee then attempts to solve the item at the improved speech-to-noise ratio.
Users can replay video clips as often as needed. If the quip still is not recognized, the speech-to-noise ratio is increased again until the trainee solves the puzzle. On the other hand, if the user has no difficulty solving an item, the speech-to-noise ratio is reduced for the next puzzle item. This way, the difficulty of the puzzle is matched to each trainee's speech recognition skills, and always poses a manageable challenge—not too difficult and not too easy.
Each trainee's performance is monitored unobtrusively, and positive feedback is provided whenever he solves an item. Scores keep track of the user’s progress, and helpful hints pop up to help the user improve speech recognition skills. (Figure 2.)
Read My Quips was developed under a grant to Advanced Hearing Concepts, Inc., from the National Institute on Deafness and Other Communication Disorders, and evaluated under a contract to the Smith-Kettlewell Eye Research Institute, which conducts research in vision and hearing loss, particularly dual sensory loss.
Nine experienced hearing aid users and one new user trained on the system during a three-week period. The trainees were provided with a laptop to use at home, and were asked to use the system whenever convenient for about 30 minutes each day, although they were allowed to use it for longer or shorter periods if they wished. A video recording of the Institute of Electrical and Electronic Engineers (IEEE) sentence test was administered in noise using an up-down adaptive paradigm to estimate the speech-to-noise ratio at a correct sentence identification rate of 50 percent. The test was administered twice to obtain an estimate of test-retest variability.
The decrease in speech-to-noise ratio at which the IEEE sentences were recognized half of the time is shown in Figure 3. The ability to understand speech at a poorer speech-to-noise ratio represents improved performance. The trainees are shown in order of the decrease in speech-to-noise ratio that they were able to handle at the end of the training program. The vertical bars show average test-retest standard error for the observed change in speech-to-noise ratio.
Subject 10 was the first-time user of acoustic amplification. She showed an improvement of 15.5 dB, which we believe was the result of auditory training and adaptation to acoustic amplification. The average improvement for the experienced hearing aid users was 2.8 dB. A statistical analysis of each subject’s performance showed that trainees 1, 2, and 3 did not show a statistically significant change in their speech-to-noise ratio for 50 percent intelligibility at the end of the training program. Six experienced hearing aid users showed a significant improvement, with an average improvement of 4.1 dB.
The time spent on the system by each trainee for each session was monitored by the computer. The average time spent on the system was 45 minutes, but there were large differences among trainees. On several occasions, the more highly motivated trainees used the system continuously for periods of up to two hours or more. Not surprisingly, the trainees showing the largest improvements in performance were the ones who spent more time using the system.
Figure 4 shows the relationship between time-on-task (the total time spent using the system over the three-week training period) and the change in speech-to-noise ratio corresponding to 50 percent intelligibility for the nine experienced hearing aid users. The correlation between time-on-task and increase in speech-to-noise ratio was 0.61.
A free trial of Read My Quips, which is available to users for $99.99, can be found on the company's website at www.sensesynergy.com. Audiologists interested in information about pricing or insurance coverage should contact Sense Synergy online.
© 2011 Lippincott Williams & Wilkins, Inc.