Not sure I fully understand the question as the speech banana is designed to show consonants and vowels (in a non scientific way) at a normal speech level on an audiogram. An example application would be to show a parent what sounds their child can hear with and without amplification to make it very clear the increased access to speech sounds. An alternative approach is to convert everything to dBSPL which you will find often called an SPL-o-gram and is often used by audiologists when evaluating hearing aid fittings using real ear measurements.

Prof Jerger recently wrote an excellent history of the audiogram which you may find helpful.

http://www.hearingreview.com/2013/03/why-the-audiogram-is-upside-down/

Regards,

Mark

This is an important fact to know as all of us are aware of the speech banana but most of us are unaware of how much is the absolute average values of speech sounds. The best way to understand that is using LTASS measures and understanding the amplitude and frequency of speech sounds used in the Speech spectrum. 

An excellent review article which explains about the origin and also provides us with the values regarding the speech spectrum is published in Ear and Hearing by Olsen, Hawkins and Van Tassel (1987). Please find the link.

Hope this information is useful

Regards,

Prashanth

http://www.ncbi.nlm.nih.gov/pubmed/3678648

Kelly, my earliest memory of the speech banana concept was with Kenneth Berger of the Kent State University in the 1970s. He called the Longterm Spectrum of Speech, from which some of the hearing aid manufacturers began shadowing the curve and range of normal speech reception onto their audiograms as a way for us to describe it to patients under examiination. If there was anyone earlier using the term I don't know who that would be. Jerger, Hawkins and others were contemporaries or slightly after Berger's work. 

Thank you for your replies. This has been helpful, especially the review article by Olsen et al 1987. I am constructing plots (audiograms) with baseline (pre-exposure) hearing levels and superimposing curves that predict hearing shifts resulting from cisplatin exposure. I want to have the 'speech banana' on the audiogram to show what speech sounds will potentially be impacted by cisplatin exposure at certain doses. I want to use this for counseling patients and as a tool for oncologists to use.

Excellent project, Kelly. We could also take your application to medicine relative to amino glycoside antibiotic ototoxicity (though delayed by about 6 months) and also the cumulative effects of NSAIs like Aleve that cause so much unattributed hearing loss. 

Thank you Max. AMGs are our next adventure. We have also seen an up tick in vancomycin use and at higher doses. We have no plans for NSAIDs at this time, but that may change in the future. Thanks for your thoughts.

Hi I'm trying to do the same as you did. In my case, I'm designing speech banana for french phonemes for counseling audiologist about percentage of errors made by humans for each phonemes in the banana speech, so it will be helpful if you tell me what you've did.

 Hi Dim, 

The project has been tabled until we get more feedback from prescribing oncologists. An early/rough draft of our idea can be seen in our publication titled "Proposed comprehensive ototoxicity monitoring program for VA healthcare (COMP-VA)", Figure 2. I think that article is available here on ResearchGate. If not, let me know and I can send it to you.

Cheers, Kelly

Hi Kelly,

Thank you for your reply,. I found the article.

Dim