Frequency Ranges
Human hearing spans roughly 20 Hz to 20 kHz, but not all frequencies are equally important — or equally reproducible by device speakers. Designing audio that works across devices means understanding these ranges and their limitations.
Human hearing range
Section titled “Human hearing range”20 Hz – 20 kHz is the textbook range, but real-world hearing varies:
- Low frequencies (bass): 20–250 Hz. Felt as much as heard. Phone speakers can’t reproduce these well.
- Mid frequencies: 250 Hz–4 kHz. Where most speech lives. Critical for intelligibility.
- High frequencies (treble): 4–20 kHz. Detail, clarity, “air.” Decreases with age.
Hearing loss typically starts with high frequencies. By age 50, many people can’t hear above 12–14 kHz.
Speech intelligibility
Section titled “Speech intelligibility”Speech fundamentally lives in the 300–3400 Hz range — which is why old telephone systems worked with such limited bandwidth. Within this:
- Fundamental frequencies: 85–255 Hz (lower for adult males, higher for children)
- Key consonants: 2–4 kHz. The “s”, “f”, “th” sounds that distinguish words
- Vowels: Lower frequencies, easier to hear
If your audio needs to communicate speech, prioritize this mid-range.
Device limitations
Section titled “Device limitations”Different devices reproduce different frequencies:
- Phone speakers: Tiny drivers struggle below 200 Hz. Heavy bass just distorts.
- Laptop speakers: Slightly better, but still limited bass.
- Earbuds/headphones: Full range, but small drivers may struggle at extremes.
- Desktop speakers: Can reproduce full range if quality is decent.
Design your audio to work on the worst-case device your users will have.
Practical guidelines
Section titled “Practical guidelines”For notification sounds
Section titled “For notification sounds”- Keep them in the 500–4000 Hz range for maximum device compatibility
- Avoid frequencies that blend into background noise
- Test on phone speakers at low volume
For voice content
Section titled “For voice content”- Ensure 300–3400 Hz is clear and uncompressed
- Use noise reduction to cut background hum (usually <200 Hz)
- Compress dynamic range so quiet speech is audible
For music and rich audio
Section titled “For music and rich audio”- Accept that it won’t sound the same everywhere
- Provide headphone recommendations for immersive experiences
- Test on multiple devices during production
Hearing accessibility
Section titled “Hearing accessibility”- Mono mixing: Some users have hearing loss in one ear
- Frequency shifting: Assistive tech can shift sounds to audible ranges
- Visual alternatives: Never rely on frequency differences alone (like high vs. low pitch alerts)
References
Section titled “References”- ITU-T G.712 — Telephony frequency bands: https://www.itu.int/rec/T-REC-G.712/en
- ASHA — Understanding Hearing Loss: https://www.asha.org/public/hearing/
- Moore, B.C.J. — An Introduction to the Psychology of Hearing: https://brill.com/display/title/33794