An audiogram is a graph that plots the softest sounds you can hear at each frequency — but the graph alone doesn’t tell the full story of your hearing. Five things matter most when reading one: the symbols and what they represent, the degree of loss at each frequency, the shape of the curve, the gap between air and bone conduction, and the pure-tone average — and understanding each one turns a confusing grid of Xs and Os into a clear picture of how your hearing actually works.
The most common mistake patients make when reading an audiogram is focusing only on degree of loss — mild, moderate, severe — and ignoring the shape of the curve. Two people can have the same pure-tone average and dramatically different hearing experiences, because the shape determines which speech sounds are missing. A sloping high-frequency loss, the most common pattern in adults, specifically removes the consonants that carry meaning — “s,” “f,” “th,” “sh,” “k” — while leaving the louder vowels intact. This is why high-frequency hearing loss produces the classic complaint: “I can hear you talking, but I can’t understand what you’re saying.”
The 5 Things That Matter on an Audiogram
The audiogram uses a specific set of symbols to plot each ear’s results separately. O (red) = right ear, air conduction. X (blue) = left ear, air conduction. < (red) = right ear, bone conduction. > (blue) = left ear, bone conduction. Air conduction tests how sound travels through the whole ear system — canal, eardrum, middle ear, and cochlea. Bone conduction bypasses the outer and middle ear and tests the cochlea directly via skull vibration. The relationship between air and bone conduction symbols is what tells your audiologist whether hearing loss originates in the outer/middle ear (conductive), the inner ear (sensorineural), or both (mixed). If you only look at the O and X — the most visible symbols — you’re seeing only half the picture.
The audiogram graph is deliberately counterintuitive: better hearing is at the top, worse hearing at the bottom. The vertical axis (y-axis) measures loudness in decibels hearing level (dB HL), running from –10 dB at the top (very soft — better than average hearing) down to 120 dB at the bottom (extremely loud). A symbol plotted high on the graph means you can hear very soft sounds at that frequency — good. A symbol plotted low means you need sounds to be very loud before you hear them — worse. The horizontal axis (x-axis) runs from low frequencies on the left (250 Hz — a deep bass hum) to high frequencies on the right (8,000 Hz — a very high-pitched whistle). Conversational speech sits primarily between 500 and 4,000 Hz. Reading the graph left to right is reading from “bass” to “treble.”
Each symbol on the audiogram marks your threshold at that frequency — the softest sound you can hear 50% of the time. The categories below reflect the WHO/ASHA degree-of-loss classification, which is the most widely used in U.S. clinical practice. The thresholds apply per frequency, not as a single number: you can have normal hearing at low frequencies and moderate loss at high frequencies simultaneously, which is exactly what the typical adult pattern looks like.
The overall pattern formed by connecting the symbols across frequencies — the “shape” of your audiogram — tells you more about your real-world hearing experience than the degree of loss at any single frequency. A sloping loss (normal at low frequencies, worse at high) is the most common adult pattern; it selectively removes consonants. A flat loss (similar thresholds across all frequencies) amplifies everything equally and responds well to hearing aids. A reverse slope (worse at low frequencies, better at high) is uncommon and produces difficulty with bass-heavy environments. A notch (isolated dip at 4,000 Hz with better hearing above and below) is the classic noise-induced hearing loss pattern. A cookie-bite (worse in the mid frequencies, better at extremes) is often genetic and causes specific difficulty with core speech frequencies.
The gap between your air conduction thresholds (O and X) and your bone conduction thresholds (< and >) at the same frequency is called the air-bone gap. A gap of more than 10 dB HL means the cochlea is hearing better than the whole ear system — which tells your audiologist that the problem lies in the outer or middle ear, not the cochlea. This is a conductive hearing loss — and it is often medically or surgically treatable. When O/X and </> symbols plot at the same level (no gap), the loss is sensorineural — originating in the cochlea or auditory nerve — and is generally not medically correctable, though hearing aids are very effective. When both a gap and reduced bone conduction are present, it is a mixed loss. The air-bone gap is the most important indicator of whether a patient needs medical evaluation before considering hearing aids.
Degree of Hearing Loss: What the dB Levels Mean
| Threshold Range | Degree of Loss | What You Typically Miss | Hearing Aid Candidacy |
|---|---|---|---|
| –10 to 20 dB HL | Normal | Nothing at this frequency — you hear soft sounds normally. | Not indicated at this threshold level alone. |
| 21 to 40 dB HL | Mild | Soft speech, whispered conversation, consonants in quiet environments. Difficulty in noise. | Often recommended, particularly when combined with high-frequency loss or speech-in-noise difficulty. |
| 41 to 55 dB HL | Moderate | Normal conversational speech at distance or in competing noise. Frequently asking for repetition. | Strongly recommended. Significant benefit expected from hearing aids across most environments. |
| 56 to 70 dB HL | Moderately severe | Most normal conversational speech without amplification. Relying heavily on lip-reading and context. | Required for functional communication. Hearing aids provide major benefit; powerful devices or custom earmolds may be needed. |
| 71 to 90 dB HL | Severe | Loud speech, environmental sounds (phone ringing, doorbell). Communication without aids is very difficult. | Power or super-power hearing aids; cochlear implant evaluation may be appropriate depending on word recognition score. |
| 91+ dB HL | Profound | Almost all sounds, including very loud environmental noise. Vibration may be perceived where sound is not. | Cochlear implant evaluation strongly recommended. Hearing aids may provide limited benefit; decision guided by word recognition score. |
Audiogram Symbol Quick Reference
Plotted in red. The most prominent symbol on most audiograms.
Plotted in blue. Paired with the O results for left-right comparison.
Plotted in red. The gap between < and O reveals loss type.
Plotted in blue. Compared to X to assess conductive component in the left ear.
Used when masking noise is introduced to the opposite ear to prevent “cross-hearing.”
Same purpose as ] but for the left ear during masked testing.
What is the pure-tone average (PTA) and why does it matter?
The pure-tone average (PTA) is a single number calculated by averaging your air conduction thresholds at 500, 1,000, and 2,000 Hz — the three frequencies at the core of the speech range. This arithmetic mean provides a rapid and simple summary of a patient’s hearing in the speech frequencies. It is the number most audiologists reference when describing degree of loss in a sentence: “You have a moderate hearing loss” means your PTA falls in the 41–55 dB HL range.
The PTA’s limitation is equally important to understand: the PTA does not provide information about high-frequency hearing loss above 2,000 Hz. Since the most common adult pattern involves a sloping loss that is worst at 4,000–8,000 Hz, the PTA can understate the real-world impact of hearing loss — particularly the difficulty with consonant clarity that drives the “I can hear but can’t understand” complaint. Your audiologist may also calculate a high-frequency PTA (average of 2,000, 4,000, and 8,000 Hz) to capture that part of the picture separately.
What is the “speech banana” and why is it on my audiogram?
The speech banana is a banana-shaped region plotted on the audiogram that represents where the sounds of normal conversational speech fall — in terms of both frequency (pitch) and intensity (loudness). Vowel sounds (A, E, I, O, U) sit in the lower frequencies, while consonant sounds are higher frequency (F, S, TH, K). When plotted together, these sounds form a curved, banana-like cluster roughly between 250–8,000 Hz and 20–60 dB HL.
The speech banana’s clinical value is immediate and visual: any of your hearing thresholds plotted below the speech banana means you are missing those sounds in conversational speech at normal speaking volumes. For a patient with a typical sloping high-frequency loss, thresholds at 2,000–8,000 Hz fall below the banana — directly in the range of the consonants “s,” “f,” “th,” “sh,” “h,” “k,” and “p” that carry the most meaning in English. This is why that patient hears voice but not words.
What does an asymmetric audiogram mean — and when should it be urgent?
An asymmetric audiogram shows a meaningful difference in thresholds between the two ears — generally defined as a difference of 15 dB or more at two or more consecutive frequencies, or 20 dB at any single frequency. Some degree of asymmetry is common and benign — prior ear infections, noise exposure in one ear, or simply individual variation. But significant asymmetry is a clinical red flag that warrants investigation before hearing aids are considered.
Conditions that can produce asymmetric hearing loss include acoustic neuroma (vestibular schwannoma) — a benign tumor on the auditory nerve — Ménière’s disease, autoimmune inner ear disease, and sudden sensorineural hearing loss (SSHL). Sudden hearing loss in one ear, particularly when accompanied by tinnitus, dizziness, or ear fullness, should be treated as a medical emergency: same-day ENT evaluation and corticosteroid treatment within 72 hours of onset offers the best chance of partial or full recovery.
At California Hearing Center, significant asymmetry on a diagnostic audiogram prompts a medical referral — we do not fit hearing aids until asymmetric loss has been evaluated for treatable causes.
What the audiogram does NOT tell you
✅ What the audiogram does tell you
- The softest sounds you can detect at each frequency — your hearing thresholds
- Whether the loss is sensorineural, conductive, or mixed (via air-bone gap)
- Which frequencies are most affected — and therefore which speech sounds you are missing
- The degree of loss at each frequency (mild through profound)
- Whether the two ears are symmetric or significantly different
- How your hearing has changed since your last test
⚠️ What the audiogram does NOT tell you
- How well you understand speech in noise — that requires a QuickSIN or equivalent speech-in-noise test
- Your word recognition score (WRS) — a separate speech test run at suprathreshold level
- Whether you have tinnitus or how severe it is
- “Percentage of hearing loss” — there is no evidence-based formula to convert the logarithmic dB scale to a percentage of hearing loss with any meaning
- The cause of your hearing loss — only the type (conductive vs. sensorineural) and configuration
- Whether you will benefit from hearing aids — that requires the full test battery and a clinical conversation
What “percentage of hearing loss” actually means: Many patients ask for their hearing loss as a percentage. This is an understandable impulse — percentages feel concrete and comparable. But the dB scale is logarithmic, not linear, and there is no clinically validated formula for converting thresholds to a percentage figure. The only valid use of “percentage” in audiology is the Workers’ Compensation disability formula — a specific legal calculation that does not translate to a meaningful description of everyday hearing ability. Your audiologist should describe your hearing in terms of degree, frequency, and configuration — not a percentage.
Why Choose California Hearing Center?
At California Hearing Center, we don’t hand you a printout and move on. Every comprehensive evaluation includes a detailed walk-through of your audiogram results — what the symbols mean, what the shape of your curve tells us about which sounds you’re missing, and what it all means for your hearing aid options and realistic outcomes. If you’ve received an audiogram elsewhere and want a second opinion or a plain-English explanation of what it shows, we’re happy to help with that too.
- Bhatt, Y. M., & de Carpentier, J. (2023). Audiogram Interpretation. In StatPearls. National Center for Biotechnology Information / NCBI Bookshelf. — Peer-reviewed clinical reference; primary source for PTA definition, frequency ranges, and audiogram notation standards. ncbi.nlm.nih.gov/books/NBK578179
- University of Iowa Carver College of Medicine. How to Read an Audiogram. Iowa Head and Neck Protocols. — Clinical protocol reference for audiogram symbols, air/bone conduction interpretation, and conductive vs. sensorineural distinction. iowaprotocols.medicine.uiowa.edu
- UCSF EARS Program. How to Read Your Audiogram. (Updated October 2025.) — Clinician-edited patient education guide; source for audiogram configuration descriptions (sloping, flat, notch, cookie-bite) and asymmetry red-flag guidance. ears.ucsf.edu
- Starkey Hearing. What Is an Audiogram and How To Read It. (2023.) — Audiogram primer including speech banana illustration and consonant/vowel frequency placement. starkey.com
- American Speech-Language-Hearing Association (ASHA). Type, Degree, and Configuration of Hearing Loss. — ASHA normative reference for degree-of-loss classification thresholds (mild 26–40 dB, moderate 41–55 dB, etc.) used throughout this article. asha.org
- Harrison, R. V. (2016). How can the audiogram be more useful? Journal of Otolaryngology — Head & Neck Surgery, 45(1). — Peer-reviewed argument for audiogram supplementation with speech-in-noise and other measures beyond pure-tone thresholds. doi:10.1186/s40463-015-0123-0