Opinion on Compression Modes

Good Morning,

Just wondering what everyone uses for their compression mode? I’m currently using the NAL mode on my Alta 2 BTE and although I like the sound, it does not seem as robust as my last pair of Oticon BTE aids(can’t remember the model).

Seems like it compresses/reduces the sound during quiet environments (like an office meeting room) and opens the sound up in louder enviroments (like in the car or at a resturaunt).

Maybe I’m just crazy, but this seems to be the opposite as my last aid. I know with the default Oticon compression there’s more fine tuning available, and so I’m curious if I’m missing new technology by not using the default Oticon setup?

Thank you!

I thought there was only different algorithms to pick up and process sound, compression is just a value that should sound the same regardless and it displayed as a muliplier. I have less than 2 on each frequency for all 3 curves…

True, I beleive you are correct that it is the algorithm. Guess I misspoke as I thought that influenced how compression functioned.

The term “compression” seems to cause a lot of confusion as it has both a technical meaning and is a commonly used word. In hearing aids, it actually has at least two technical meanings from what I can tell. When used alone, it refers to how much gain sounds of different loudness get. Soft sounds usually get substantially more gain than loud sounds so that the soft sounds are audible and the loud sounds don’t overwhelm. Hence the loudness levels are “compressed” from what is “normal.” A loud sound will not be proportionally louder than a soft sound when processed through the hearing aid. “Frequency Compression” is an entirely different thing and is a type of Frequency Lowering. This was meant to clarify. If it confuses, kindly ignore.

If you’re talking about NAL mode, you’re probably talking about Fitting Rationale that HA mfgs use as a basis for their amplification. Some of the industry standard fitting rationales are NAL-NL1, NAL-NL2, DSL i/o, etc. Sometimes HA mfg may use their own proprietary fitting rationale that they deem best fit for their devices, like Oticon has what they call VAC+ as their own proprietary fitting rationale. I would have assumed that the Alta 2 BTE, being an Oticon brand HA, would have used the VAC+ instead of one of the NAL rationales, unless your audi switch it to one of the NALs for you for some reason. If you like the fitting rationale of your previous pair of Oticon BTE, there’s a good chance that it was the VAC+, so you may want to add your audi to switch you back to the VAC+ to see if you like that better or not.

Below is a description of what a fitting rationale is by Donald Schum: A fitting rationale is a set of formulas used to fit a broad range of sounds in the environment (usually speech) into the remaining dynamic range of the patient. It will take into account factors such as the patient’s thresholds, the measured or predicted UCL, the importance factor of the various speech frequencies, sound quality concerns and psychoacoustic dimensions such as loudness perception.

As for compression, it can imply Wide Dynamic Range Compression (WDRC), which is what MDB explained to you, compressing a wide range of sound intensities into the limited dynamic range of the hearing aid user. For example, sound intensities usually range between 10 and 100dB, from very soft sounds to very loud sounds that can be comfortably be heard by a normal hearing person. That’s basically a 90db dynamic range. But for a hearing challenged person who has a loss of say 40dB (meaning that the softest sound they can hear is at 40dB), then their dynamic range is only 60dB (100dB minus 40dB). So a normal soft sound at 10dB will need to be amplified to 40dB for that person to hear. But you don’t want to amplify a 100dB sound to 140dB for that person to hear because that would be too loud, and that person can hear the 100dB sound just fine. So as the normal sound progressively gets louder, the amplification amount gets progressively less, until 100dB sounds need no amplification at all. That’s WDRC -> the wide dynamic range of 90db gets compressed to a narrower dynamic range of 60dB to fit into this hearing aid user’s more limited audible dynamic range. Of course this is adjustment is done differently in multiple frequency bands depending on what the hearing loss is in each frequency band. The more bands, the more fine tuned to the individual hearing loss the adjustment can be.

Another type of compression is the frequency lowering that MDB also mentioned. But this is frequency compression and not volume/loudness compression.

A third type of compression is for when the input sound is too loud, louder than the input mic and the limitation of the input dynamic range of the hearing aid system can handle. For example, a 16-bit analog to digital converter in a hearing aid system only has a theoretical dynamic range of 96dB. So any loud sound outside of this 96dB input dynamic range will get compressed so that it will fit into the 96dB range or else the sound will get clipped and sound saturated/distorted.

1 Like

Woah, thank you and MDB for those very in-depth explinations. There’s apparently a lot more nuance to the word “compression” than I thought in regards to aids.

For fitting rationales, is there an advantage to one or another? I know I’m using the NAL one, which seems to have a much more limited customization ability compared to the VAC+ (for example, there’s whole options such as speed of compression/gain that are not accesible with the NAL fitting)

I should probably have my settings re-done with the VAC+ option and see if that improves my issues.

Thank you again for the in-depth posts!

Customization ability may have to do with what brand’s software is being used. In Connexx 8, when I use the NAL 2 fitting, I have a ton of options available under the Parameters tab. From my understanding, which fitting formula is used is pretty much personal preference. From reading, it is my understanding that manufacturer’s fitting formulas tend to emphasize comfort versus NAL and DSL tend to emphasize audibility.