How to Make a Telephone or Radio Vocal Effect in Your DAW

What makes a vocal sound like a telephone or radio?

A telephone or radio vocal is mostly missing frequencies. Classic phone and radio systems passed only a slice of the audio spectrum, so the effect starts with cutting lows and highs, then adds distortion, small-speaker coloration and heavy compression.

The numbers come from real systems. The G.711 telephone codec passes audio in the 300-3400 Hz band and samples it at 8000 Hz with 8 bits per sample^[1]. AM radio in the US is a little wider: in 1986 the National Radio Systems Committee set an AM audio bandwidth limit of 10 kHz^[2], and cheap receivers and speakers often reproduce less than that. Your ear knows these bandwidths from decades of calls and car radios, which is why a few EQ moves are so convincing.

Treat the table as starting points and tune by ear. Only the telephone and AM figures come from the standards above; the rest are practical ranges for the character.

What EQ settings make a telephone vocal?

Use a steep high-pass filter around 300 Hz and a steep low-pass around 3-3.5 kHz, then add a narrow boost somewhere between 1 and 2 kHz for the honky, resonant quality of a small earpiece. The steepness is what sells it; gentle slopes just sound like a dull vocal.

Most stock EQs can do this. Ableton's EQ Eight offers low cut and high cut filters at either 12 or 48 dB per octave^[3], and the 48 dB setting is the one you want here. If you prefer one knob, Ableton's Auto Filter includes a band-pass type^[3], which is handy for sweeping the effect later. For the basics of filter types and slopes, see what is EQ.

Sweep the resonant boost while the vocal plays. Around 1 kHz sounds boxy and nasal, closer to 2 kHz sounds more cutting and modern. Pick the spot that suits the song rather than a fixed number.

How much distortion should a radio or telephone vocal have?

Use enough distortion to add grit on the loud words while the quiet ones stay clean. For most telephone effects that means light saturation; walkie-talkie and megaphone sounds can take hard clipping.

Order matters. Distortion generates new harmonics, some above your low-pass filter, so either place the distortion before the EQ or add a second band-pass after it. A common chain is EQ, then distortion, then a second EQ to trim the fizz the distortion created. The differences between saturation, overdrive and fuzz are laid out in saturation vs distortion vs overdrive vs fuzz.

• Telephone Soft saturation or tape drive. Aim for grit you notice only when you bypass it.
• AM radio Tube-style saturation plus a compressor pushed hard, which mimics broadcast limiting.
• Walkie-talkie or megaphone Hard clipping or amp simulation, with the band narrowed further to roughly 500 Hz to 3 kHz.
• Parallel option If the words get lost, blend a heavily distorted copy under a lightly filtered one on a bus.

Heavy compression belongs in this chain too. Real phone lines and broadcast chains flatten dynamics, so fast attack, fast release and 6-10 dB of gain reduction are a reasonable starting point. If you need a free drive plugin for this, the best free saturation plugins covers several.

Should you bitcrush a telephone vocal?

A little bitcrushing adds the digital harshness of a mobile call or a cheap intercom, but it is optional. Use it for a modern, glitchy phone sound; skip it for a warm vintage radio.

Apple describes Logic Pro's Bitcrusher as a low-resolution digital distortion that can emulate early digital devices and create artificial aliasing. Its Resolution control ranges from 1 to 24 bits, and its Downsampling control divides the sample rate, so 10x turns a 44.1 kHz signal into 4.41 kHz without changing playback speed or pitch^[4].

Place the bitcrusher before your band-pass EQ, so the filter tames the harsh aliasing above the telephone band. Dedicated third-party crushers go further than stock tools if you want more character.

How do you simulate a small speaker or phone speaker?

Small speakers have no real bass, a peaky midrange and some distortion when pushed. You can fake that with EQ alone, or more convincingly with an impulse response of a real small speaker loaded into a convolution plugin.

The EQ version uses the phone-speaker row from the table above: a high-pass around 400 Hz, one or two resonant peaks between 1 and 4 kHz, and a gentle roll-off above about 8 kHz. For the convolution version, load an impulse response of a phone, laptop or transistor radio speaker at 100% wet; free IR collections are compared in free convolution reverb and IR packs. Listening to real mixes on a phone, as described in the phone speaker mix test, is the fastest way to learn what the target sounds like.

• Keep it mono Phones and old radios are mono sources. Pan the effect vocal center or place it as one mono point in the stereo field.
• Add room or noise A short room reverb or a faint noise bed under the vocal makes it sound like it is coming out of a device, not just filtered.
• Cut the reverb before the filter If you use a reverb or delay send, filter the return too, or the full-range reverb will give the trick away.

What is a good telephone vocal chain, step by step?

Build the effect on its own bus or return track, in this order, so you can switch it on and off without touching the main vocal chain. Adjust each step by ear against the beat.

1. Create an effect bus
   Duplicate the vocal or send it to a new bus named PHONE. The dry vocal chain stays untouched.
2. Optional bitcrusher
   10-12 bits, light downsampling, 20-50% mix.
3. Band-pass EQ
   48 dB/oct high-pass at 300-400 Hz and low-pass at 3-3.5 kHz, plus a narrow 1-2 kHz boost.
4. Distortion
   Light saturation for phone, harder clipping for megaphone. Watch that consonants stay clear.
5. Second EQ
   Repeat the band-pass to remove the fizz and rumble created by the distortion.
6. Compressor
   Fast attack and release, 6-10 dB of gain reduction for a broadcast-flat sound.
7. Level match
   Filtering removes energy, so raise the bus until the effect sits at the same perceived loudness as the normal vocal.

How do you automate the effect for intros and bridges?

Automate the switch between the normal vocal and the effect bus, or automate the filter cutoffs so the effect opens up into the next section. The contrast is the point: a telephone intro makes the full-range vocal hit harder when the drop arrives.

In Ableton Live, turn on Automation Arm and every control change you make while Arrangement Record is running becomes automation^[5], so you can perform the sweep live and clean it up afterward. Drawing the envelopes by hand works just as well; the general workflow is in the DAW automation guide.

When you open the filters into a chorus, cut the effect a few milliseconds before the downbeat so the full vocal lands exactly on it. And check the transition on small speakers, since the effect vocal can seem louder there than on your monitors.