Spectrogram Viewer
Visualize how frequency content changes over time from an audio file or live microphone input.
Microphone audio is analyzed locally in your browser. Permission is requested only after you start.
How to read it: time runs left to right, frequency runs bottom to top, and brighter regions indicate stronger spectral energy.
Accuracy note: FFT size, sample rate, source quality, microphone response, and sensitivity settings affect how the visualization appears.
Privacy: this tool processes selected audio or microphone input locally in your browser and does not intentionally upload the analyzed audio.
A spectrogram viewer shows how the frequency content of a sound changes over time. Upload an audio file or use your microphone, and the tool converts the sound into a visual time-frequency display where position shows time and frequency while color intensity represents spectral energy.
This is useful for examining vocals, instruments, harmonics, background noise, resonances, transients, and other details that are difficult to understand from a waveform alone. For a deeper walkthrough of the visual format, read how to read a spectrogram.
How to Use the Spectrogram Viewer
You can analyze sound in two ways.
Upload Audio: drag an audio file into the tool or browse from your device.
Live Microphone: start microphone access and watch the spectrogram update in real time.
After choosing a source, you can adjust the FFT size, maximum displayed frequency, frequency scale, and sensitivity. The spectrogram can also be paused, cleared, or exported as an image.
A good starting setup is:
- FFT size: 2048
- Frequency range: up to 10 kHz or 20 kHz
- Linear or logarithmic scale depending on the task
- Moderate sensitivity
If you are analyzing a recording with strong background interference, see the guide on noise and background interference for practical context.
How to Read a Spectrogram
A spectrogram has three main dimensions.
Horizontal axis — Time
Moving from left to right shows how the sound changes as the recording progresses.
Vertical axis — Frequency
Low frequencies appear toward the lower part of the visualization, while higher frequencies appear farther up.
Color or brightness — Intensity
Brighter or stronger regions indicate greater spectral energy at that time and frequency.
For example, a sustained musical note may produce a series of relatively horizontal bands. A short drum hit often appears as a brief vertical burst because its energy spreads across many frequencies in a short period.
Understanding this relationship makes a spectrogram much more useful than simply looking at the colors.
Spectrogram vs Spectrum Analyzer
A spectrogram and a spectrum analyzer are closely related, but they answer different questions.
A spectrum analyzer shows how much energy exists at different frequencies at a particular moment or over an averaged interval.
A spectrogram adds time as another dimension.
That means a spectrum can tell you that strong energy exists near 440 Hz, while a spectrogram can show exactly when that 440 Hz energy appears, how long it lasts, and what other frequencies appear alongside it.
This time-frequency relationship is why spectrograms are useful for studying changing sounds such as speech, singing, instruments, and music.
For more background on frequency itself, read what is the difference between pitch and frequency.
Fundamental Frequency and Harmonics
Many musical sounds contain more than one frequency.
When you sing or play a pitched instrument, the fundamental frequency usually corresponds closely to the perceived note. Additional frequencies called harmonics occur above the fundamental.
For example, if the fundamental is 220 Hz, related harmonic energy may appear near:
- 440 Hz
- 660 Hz
- 880 Hz
- higher multiples
On a spectrogram, these can appear as stacked horizontal bands.
The relative strength of those harmonics contributes to timbre, which is one reason a guitar and a human voice can sound different even when both produce the same musical note.
If a tuner or detector seems to react to multiple frequencies, the guide why tuners show multiple notes: harmonics explained explains this behavior in more detail.
What FFT Size Means
The spectrogram uses frequency analysis based on the Fast Fourier Transform, or FFT.
FFT size affects the balance between time resolution and frequency resolution.
A smaller FFT size responds more quickly to short changes but provides less precise separation between nearby frequencies.
A larger FFT size provides finer frequency detail but spreads analysis across a longer time window.
For example:
- 512: good for fast transients
- 1024: balanced for general visualization
- 2048: useful for voice and musical analysis
- 4096: better frequency separation but less precise timing detail
There is no universally best FFT size. The right choice depends on what you want to inspect.
For a technical explanation of the underlying process, see how FFT works in pitch detection.
Linear vs Logarithmic Frequency Scale
The viewer can display frequency using either a linear or logarithmic scale.
With a linear scale, equal distances represent equal numbers of hertz. The space between 100 Hz and 200 Hz is therefore visually similar to the space between 1,000 Hz and 1,100 Hz.
A logarithmic scale is often more intuitive for music because musical intervals are based on frequency ratios.
For example:
- 110 Hz to 220 Hz = one octave
- 220 Hz to 440 Hz = one octave
- 440 Hz to 880 Hz = one octave
Even though the number of hertz doubles each time, the musical interval is identical.
For more on the connection between musical pitch, octave, and frequency, read frequency vs note vs octave.
Using a Spectrogram for Voice Analysis
Voice spectrograms can reveal several useful features.
You may see:
- a fundamental pitch
- harmonic bands
- changing pitch over time
- vibrato patterns
- consonant transients
- breath noise
- resonant frequency regions
A steady sung vowel often produces clearer horizontal harmonic bands than normal speech because the pitch remains more stable.
The spectrogram can therefore be useful for learning how voice frequency changes, but it does not automatically diagnose vocal technique or health.
For a broader reference covering typical sound ranges, see frequency ranges for instruments and voices.
Using Spectrograms for Instruments and Music Production
Musicians and producers can use spectrograms to inspect:
- low-frequency rumble
- resonances
- harmonic content
- cymbal energy
- vocal sibilance
- transients
- sustained notes
- broadband noise
- overlapping instruments
For example, a bass instrument may show concentrated energy in the lower part of the spectrogram with harmonics extending upward. Cymbals tend to produce broader high-frequency energy.
This makes the visualization useful for understanding why two sounds occupy similar or different frequency regions.
Audio engineers who want more context can read pitch detection tools for audio engineers.
Maximum Frequency and Sample Rate
The highest meaningful frequency that can be represented depends partly on the audio sample rate.
Digital audio is limited by the Nyquist relationship: the highest representable frequency is approximately half the sample rate.
For example, audio sampled at 48 kHz has a theoretical Nyquist frequency of 24 kHz.
The viewer lets you choose a lower maximum display frequency when you want to focus on a specific area such as vocals or low-frequency instruments.
Showing a smaller range can make important details easier to see.
Microphone and Recording Quality
Live microphone spectrograms depend heavily on the incoming signal.
Results can be affected by:
- microphone frequency response
- room reflections
- background noise
- automatic gain control
- device processing
- microphone distance
- clipping
- environmental sounds
A weak microphone may not capture very low or high frequencies accurately, while aggressive device processing can change how the spectrum appears.
The article how microphone quality affects pitch detection accuracy explains why the input device matters for frequency-based analysis.
Accuracy and Limitations
A spectrogram should be understood as a visual analysis tool rather than an exact measurement of everything in the sound.
The appearance can change depending on:
- FFT size
- sensitivity
- frequency scale
- sample rate
- browser performance
- source quality
- microphone quality
- background noise
Strong colors do not directly represent perceived loudness, and a visible frequency component does not automatically tell you which instrument or sound produced it.
The tool also does not automatically identify every musical note, chord, singer, or instrument.
Privacy
Uploaded audio is processed locally by the tool in your browser.
Microphone access is requested only when you start the live mode. The plugin is not designed to intentionally upload or record microphone audio, and microphone tracks are stopped when live analysis ends.
You can also clear the current visualization or reset the tool when you are finished.
Common Questions
What is a spectrogram?
A spectrogram is a visual representation of frequency content over time.
What do the colors mean?
They represent relative spectral intensity. Stronger colors generally indicate more energy at that time and frequency.
Can a spectrogram show musical notes?
It can show frequency patterns associated with notes, including fundamentals and harmonics, but it does not automatically label every visible component as a musical note.
Why do I see several horizontal lines when singing one note?
Those lines are often the fundamental frequency and its harmonics.
What FFT size should I use?
Start with 2048. Use a smaller FFT for better timing detail or a larger FFT for finer frequency separation.
Can I analyze microphone input live?
Yes. The tool includes a live microphone mode.
Can I save the visualization?
Yes. The spectrogram can be exported as an image.
Why does the same audio look different after changing settings?
FFT size, frequency scale, maximum frequency, and sensitivity all affect how spectral information is displayed.
Use the Spectrogram to See What Your Ears Hear
A spectrogram turns changing sound into a visual pattern. Use it to inspect where frequency energy occurs, identify harmonic structures, compare sustained and transient sounds, or explore how vocals and instruments occupy the frequency spectrum.
For deeper learning, continue with how to read a spectrogram or compare the visualization with the site’s Frequency Detector when you need a simpler live frequency-focused result.