Voice Amplifier on iPhone: Gain Without Feedback
A voice amplifier app on an iPhone makes you louder by taking the phone's own microphone and sending it live to a speaker iOS is already playing through: a Bluetooth speaker, a soundbar, a car stereo, AirPlay, or a wired connection. In a living room, a classroom or a backyard, that is genuinely enough. It is not a hearing aid, it is not an assistive listening device, and it will not fill a hall of two hundred people.
What a voice amplifier app on an iPhone actually does
The phone is the microphone. Every limit below falls out of that. The mic is read in 2048-sample buffers, fed through a gain stage you control from 0 to 200 percent, then an optional echo-reduction high-pass filter, then the filter stages of the voice preset you picked, then a level meter, then out to whatever route iOS has already selected.
That last step is the one people get wrong. The app does not pair anything. No iPhone app can pair a Bluetooth device, because iOS exposes no public API for it; pairing happens in Settings and nowhere else. The app plays into the route iOS hands it. If your speaker is connected and music comes out of it, your voice comes out of it too. If nothing is connected, it comes out of the phone's own small speaker, the worst option and the one most likely to howl.
Android differs on paper and matches in practice. Android has handed apps pairing APIs for years, BluetoothDevice.createBond since API 19 and CompanionDeviceManager since API 26, so an Android app is technically allowed to bond a speaker itself. A mic-to-speaker app still plays into the output the system has already selected, so the steps are the same on both platforms: connect the speaker in the phone's own Bluetooth settings, confirm music comes out of it, then open the app.
You can hear the entire chain before installing anything. The live voice effects tool runs the same eight presets in a browser tab and draws the real frequency response under each one.
Nothing is recorded, nothing is uploaded, and there is no server. The audio goes mic to speaker and stops existing.
How loud it gets, and which rooms that covers
The ceiling is the speaker, not the app. Gain past 100 percent buys headroom for a quiet talker or a phone held at arm's length; it does not add watts to a speaker that has none. To amplify your voice on an iPhone in a bigger space, the honest upgrade is a bigger speaker, not a better app.
| Space | What you are working with | Verdict |
|---|---|---|
| Living room or dinner table, up to about 15 people | One mid-size Bluetooth speaker | Works well |
| Quiet classroom or meeting room, 25-30 seated | Speaker at the front, aimed at the seats | Works |
| Backyard or driveway, 20-40 people, music paused | A party speaker with real output | Works for announcements |
| Hall, gym or church, 100+ people | Anything consumer | No. Rent a PA |
Check what your voice looks like at the distance you will really hold the phone. The browser mic monitor shows the input level while you talk, which beats finding out in front of an audience.
When the live output route is the iPhone's own speaker or earpiece, the app caps gain at 0.6 and warns you: a mic and a speaker four centimetres apart is a feedback loop by construction. The cap lifts once Bluetooth, headphones or AirPlay takes the route. If it will not get louder, check what it is playing through.
The megaphone preset, and the other seven
People search for a megaphone app for iPhone expecting a novelty. This one is closer to a tool, because its filter shape is what a real megaphone does: a high-pass at 600 Hz Q 0.8 to throw away everything below the voice, a bandpass at 1.5 kHz Q 2.5, and a +12 dB peaking boost at 2.5 kHz Q 3. That sits on the consonant range, where speech intelligibility mostly lives. Stripping everything under 600 Hz also removes low-mid energy a small room bounces back into the microphone.
| Preset | Filters | Use it for |
|---|---|---|
| Normal | None | Straight amplification |
| Megaphone | Highpass 600 Hz Q0.8, bandpass 1.5 kHz Q2.5, peaking 2.5 kHz Q3 +12 dB | Cutting through a noisy room |
| Bright | Highshelf 3.5 kHz +10 dB, peaking 6 kHz Q1 +4 dB | A dull speaker |
| Radio | Highpass 500 Hz Q0.7, bandpass 1.8 kHz Q5, lowpass 3.2 kHz Q0.7 | Announcements |
| Deep | Lowpass 480 Hz Q0.9, lowshelf 160 Hz +6 dB | Fun, bad for speech |
| Warm | Lowshelf 240 Hz +8 dB, lowpass 6.5 kHz Q0.7 | Harsh, thin speakers |
| Space | Bandpass 950 Hz Q1.1, peaking 1.6 kHz Q1 +4 dB | Fun |
| Robot | Bandpass 1.2 kHz Q1.2, peaking 1.8 kHz Q1 +5 dB, 50 Hz ring modulator | Fun |
For real amplification you want Normal, Megaphone or Bright. The other five are for kids and parties. Compare all eight against your own voice on the voice effects page; the curve under each shows why Deep makes you unintelligible once a room gets loud.
What it is not: hearing help, and not a real PA
This matters more than the feature list. A phone plus a speaker is a small PA system. It is not an assistive listening device and not a hearing aid, it was never designed or tested as one, and nobody should treat it as one. If someone is struggling to follow conversation, that is a question for an audiologist.
The related confusion is Apple's Live Listen, which routes the iPhone microphone to AirPods or a compatible hearing device. It does not send that audio to a Bluetooth speaker, which is why people go looking for a third-party app. Why Live Listen will not talk to a Bluetooth speaker covers the rest.
One direction does not exist: the iPhone is the microphone, never the receiver. It cannot play audio coming from another device's mic. Audio flows out, not in.
A real PA wins for a concrete reason. A handheld or lavalier mic sits centimetres from your mouth on a wire: huge signal, little gain, almost none of the room. A phone on a table hears everything.
The delay you will hear
Bluetooth adds latency, and no app removes it, because it lives in the codec and the receiving speaker's buffer rather than in software on the phone. No app can hand you a zero-latency link over Bluetooth. The published figures genuinely disagree with each other:
- A2DP end to end: roughly 100-300 ms (Hollyland).
- SBC: quoted as 100-150 ms by some sources and 150-300 ms by others. Both are out there.
- AAC: roughly 100-200 ms. This is the codec an iPhone uses.
- aptX Low Latency: about 40 ms (Qualcomm). An iPhone never negotiates it, because iPhone has no aptX support of any kind; it sends AAC.
- LC3 / LE Audio: about 20-30 ms (Bluetooth SIG).
- Dedicated 2.4 GHz wireless microphone systems: under 20 ms.
Under about 20 ms goes unnoticed, 40-50 ms is where lips stop matching sound on video, and past roughly 100 ms it feels like a bad phone call. Self-monitoring is far more sensitive than passive listening, because your skull delivers the original instantly and the speaker delivers the copy a beat later. The perception thresholds are covered here, and the consequences for a live mic in the guide on Bluetooth microphone delay.
An Android phone has one lever an iPhone does not. Since Android 8.0 the platform can negotiate AAC, aptX, aptX HD and LDAC on top of mandatory SBC, and Developer options exposes a codec picker that lets you force one of them. Only LDAC's encoder ships in AOSP; aptX and aptX HD are proprietary Qualcomm code each manufacturer licenses separately, so what your handset offers depends on the handset. Changing codec moves the delay somewhat. It does not transform it, and iOS has no equivalent setting at all.
Do not take anyone's numbers, including these. The latency test plays a 30 ms chirp, records it back through your mic, cross-correlates five runs and subtracts air time for the distance you enter. That median is your speaker's number, the only one that matters. If it comes back ugly, a wired connection closes the gap for a few dollars.
Feedback is what actually ruins it
Amplified sound reaches the microphone, gets amplified again, and the loop runs away. That squeal is the main reason a phone-as-PA setup fails in front of people, and it is pure geometry.
The app has two defences: the gain cap described above, and an echo-reduction high-pass filter that sweeps upward from 50 Hz, sitting at 50 + 110 times the level setting in hertz with Q 0.8, while switching the iOS audio session into voiceChat mode. That removes low-frequency energy and enables the system's own echo handling.
Geometry beats settings. Stand behind or beside the speaker, never in front. Aim it at the audience, away from the phone. Keep distance between the two, because halving that distance roughly quadruples the energy arriving back at the mic. Then set gain to the lowest level that works, not the highest that has not squealed.
If you are already hearing it, the guide to microphone feedback with a Bluetooth speaker covers diagnosing which frequency is looping.
Setting it up before people are watching
- Pair the speaker in iOS Settings first. No iPhone app can do it for you, this one included.
- Play a song and confirm it comes from the speaker, not the phone. People skip this.
- Open the app, start it, set gain around 60 percent, and speak normally.
- Walk to the back of the room while someone else talks into the phone. Judge it from there.
- Turn on echo reduction at any hint of ringing, and pick Megaphone or Bright in a noisy room.
- Turn on Do Not Disturb so a notification tone does not land in the middle of a sentence.
The connection itself is covered step by step in connecting an iPhone microphone to a Bluetooth speaker. For an event, run the pre-party microphone checklist the day before, not an hour before. For repeated announcements rather than a speech, the announcements guide covers that setup, battery included.
Bluetooth mic: Mic to Speaker is free on both stores. The App Store build needs iOS 15.1 or later and runs on iPhone and iPad; the same app is on Google Play from AppsOverflow LLC, so an Android reader who landed here can install it too. No account, no ads and nothing recorded on either platform. To try the idea first, the browser tools on this site run the same processing for the price of a microphone permission.
Try the presets on your own voice
All eight of the app’s voice presets, live on your own voice, with the frequency-response curve each one applies.