How to Use Spectrograms to Learn Details of Bird Songs

Learning Songs & Calls | Basic Parts of a Song | Spectrograms | Mnemonics

Bird songs and calls can be hard to learn—they’re so intricate, and they go by so fast. That’s why ornithologists use spectrograms—visualizations that allow you to see the shape of a song, freeze it in time, and study it in detail. Using a simple graph format, they show a sound’s frequency (or pitch) and loudness over time. You can use your eyes as well as your ears to pick out subtle features of the sound—details you can hear next time you listen to it.

How to Read a Spectrogram

A spectrogram shows sounds much the same way a sheet of music does: You read it from left to right, and pitch is on the vertical axis. Higher-pitched sounds appear higher on the graph; deep notes are toward the bottom. The heavier the mark, the louder the sound is at that moment. The thicker the line, the coarser or buzzier the sound is.

As an aid to listeners, our Macaulay Library displays spectrograms for each one of the 3+ million sounds in its archive. When you tap the play arrow, the sound plays and the spectrogram scrolls past, showing you where you are in the song. You can pause or go back at any time. We’ve collected a few examples of simple sounds here, to get you started.

Whistles

When learning to read spectrograms, it helps to start with a simple sound like a whistle. This Harris’s Sparrow sings a pure, whistled song on one pitch. Its spectrogram looks like this:

Harris’s Sparrow song. The flat lines on the spectrogram indicate clear notes that stay on the same pitch. Each line is a note, and the gap between the two lines is a short break in the song.

If you keep listening, you’ll see the bird’s next song scroll past at the 6-second mark. This time you can see there are three notes, and the lines are lower in the spectrogram—these notes are lower-pitched than in the first song.

---

White-throated Sparrows sing a similar whistled song, but the gaps between the notes are a bit longer and they fall in pitch as the song goes on. You can see these details in the spectrogram:

White-throated Sparrow song. The lowest lines in the spectrogram are the darkest, indicating these are the loudest parts of the notes. The lighter lines above them are quieter sounds known as harmonics or overtones (see below).

---

Northern Cardinals also whistle, but there’s more going on in their songs. Their notes rise and fall sharply, and the rhythm changes.

Northern Cardinal song. The first few notes are long, often with a “laser gun” quality as the notes slur upwards or downwards. Cardinal songs often end with a series of shorter, simpler notes. By looking at the spectrogram, you can see how these short notes begin with a quick rise in pitch before leveling off, creating an abrupt sound.

Buzzes

Whereas thin lines on a spectrogram indicate clear whistles, thick and blurry lines means the sound is incorporating multiple pitches at the same time. We hear this as a buzzy tone. 

Golden-winged Warbler song. The general pattern of this song is similar to White-throated Sparrow—a long introductory note followed by additional notes at a lower pitch. But the thick lines indicate the notes will sound buzzy. 

Chips

When seen in spectrograms, chip notes are sort of the opposite of whistles. Where a whistle is a thin horizontal line representing a long, clear tone, a chip is a very short note covering a wide range of pitches. Chips appear as vertical lines in a spectrogram.

Worm-eating Warbler chip notes. Each of these vertical lines represents a single “chip” note.

Trills

Putting chip notes together into a long series of very fast notes creates a trill. By clearly showing how closely spaced the notes are, spectrograms make it easy to appreciate the relative speed of a bird’s trill.

Fast trill: Worm-eating Warbler. Each of the vertical lines represents a single chip note. This individual sings at about 29 notes per second.

Slower trill: Pine Warbler. This Pine Warbler has a more relaxed sound to its trill—caused by singing only about 13 notes per second. 

Harmonics

Often on spectrograms you’ll notice fainter “copies” of notes at higher (or lower) pitches than the main part of a song. These are known as harmonics, and they can lend an extra quality to the sound—sometimes making whistles sound richer; other times making the sound distinctly nasal. Some species, like Red-breasted Nuthatch, produce many prominent harmonics, which contribute to the characteristic nasal tone of their songs and calls. 

Red-breasted Nuthatch call. The insistent “yank-yank” call of the Red-breasted Nuthatch has a nasal tone thanks to the many harmonics in the sound.

Warbling

The beautiful, warbling songs of American Robins and other songbirds can contain many syllables, sudden changes in pitch, multiple harmonics, and more. Spectrograms help appreciate the number and distinctiveness of each syllable—and they can also give a clue to subtle differences in tone.

Compare these American Robin and Rose-breasted Grosbeak songs—two similar songs that can be confusing in the field. On the spectrograms, the robin’s notes show up as thicker, more diffuse lines on the spectrogram whereas the grosbeak’s notes are drawn more sharply. This is a graphic representation of the key difference: the robin has a hoarser tone compared with the sweetly whistled sound of the grosbeak. (In fact, grosbeaks are sometimes described as “a robin that’s had voice lessons.”)

Complex Songs

Many birds sing long, varied songs that mix together whistles, chips, trills, and other sounds. Spectrograms can be incredibly helpful in breaking down such songs so you can understand what you’re hearing—and have a better chance of recalling it.  

A whistle that turns into a trill: Black-chinned Sparrow. This desert bird sings an accelerating song. It starts with one simple, high-pitched whistle, then switches to several quick rising whistles that gradually turn into a tight trill by the end.

---

Winter Wren song. Tiny, mouse-brown Winter Wrens sing some of the most complex songs of any bird. Their songs are loud, long, and switch between whistles, chips, slurs, and trills in a split-second. The spectrogram helps pause the action, break the song apart, and recognize the many different elements.