Nonlinear Frequency Compression Algorithm As conventional amplification does not provide sufficient high-frequency information, an alternative approach is provided by frequency-lowering. This can be achieved by different methods. While all shift the high frequencies to a lower region there are large differences in implementation, which can significantly affect sound quality, speech intelligibility, and awareness of ... Inbox
Free
Inbox  |   March 01, 2009
Nonlinear Frequency Compression Algorithm
Author Notes
  • Myriel Nyffeler, Stäfa, Switzerland Myriel.Nyffeler@phonak.co
    Myriel Nyffeler, Stäfa, Switzerland Myriel.Nyffeler@phonak.co×
  • Editor's note: The author is Phonak's international field study coordinator.
    Editor's note: The author is Phonak's international field study coordinator.×
Article Information
Speech, Voice & Prosodic Disorders / Research Issues, Methods & Evidence-Based Practice / Speech, Voice & Prosody / Inbox
Inbox   |   March 01, 2009
Nonlinear Frequency Compression Algorithm
The ASHA Leader, March 2009, Vol. 14, 4-46. doi:10.1044/leader.IN3.14042009.4
The ASHA Leader, March 2009, Vol. 14, 4-46. doi:10.1044/leader.IN3.14042009.4
As conventional amplification does not provide sufficient high-frequency information, an alternative approach is provided by frequency-lowering. This can be achieved by different methods. While all shift the high frequencies to a lower region there are large differences in implementation, which can significantly affect sound quality, speech intelligibility, and awareness of environmental sounds.
Francis Kuk et al. describe a method of frequency transposition (“Re-evaluating the Efficacy of Frequency Transposition,” Jan. 20, 2009). They also claim that it is better than frequency compression because “In frequency compression…sounds in the aidable region from 0 Hz to 4000 Hz which do not need to be lowered are nonetheless compressed.” Although this may be a correct description of a particular implementation of frequency compression, it leaves the misleading impression that all frequency compression algorithms are implemented in this way.
The SoundRecover algorithm, available from Phonak, compresses frequencies above a programmable cut-off frequency and shifts them to lower frequencies, whereas the frequencies below the cut-off are amplified normally. Although input frequencies up to a defined cut-off frequency do not undergo any frequency compression, nonlinear frequency compression is applied in increasing degrees to higher input frequencies. The figure illustrates the effect of changing the cut-off frequency with (a) showing a high cut-off frequency (at 3 kHz) and a high compression ratio and (b) a lower cut-off frequency (at 1.5 kHz) and a lower compression ratio. Study results have shown no difference in vowel recognition with frequency compression while high frequency consonant recognition has significantly improved with SoundRecover.
0 Comments
Submit a Comment
Submit A Comment
Name
Comment Title
Comment


This feature is available to Subscribers Only
Sign In or Create an Account ×
FROM THIS ISSUE
March 2009
Volume 14, Issue 4