How to warm up thin or tinny voice recordings
- Step 1Make sure you're on Pro — Voice EQ requires Pro (
minTier: pro). On Free it's locked and the audio family caps at 50 MB / 30 min. Pro gives 200 MB / 120 min per file. - Step 2Drop the thin recording in — One file per job. MP3, WAV, FLAC, M4A, OGG, and Opus decode. A clean single-voice track warms up best.
- Step 3Select the Warm profile — Choose Warm from the Voice profile picker. It's the only one tuned to preserve body: 70 Hz high-pass, -1 dB @ 350 Hz, +2 dB @ 2.5 kHz. If the voice still feels small, A/B it against Male to hear the difference the gentler curve makes.
- Step 4Process locally — FFmpeg runs the chain as a single
-afpass in your browser. Seconds to under a minute for a typical track; the first run loads the WASM engine once. - Step 5Compare against the raw file — A/B Warm against the original. You're listening for a fuller, rounder body without new boominess. If it's now boomy, the source already had enough low end — try Male instead.
- Step 6If it's still harsh up top, de-ess — Warm is the gentlest profile, but a tinny mic can still be sibilant. Run the de-esser after this tool to soften harsh S sounds without removing the warmth.
Warm profile vs the other three
Why Warm sounds warmer: it removes and cuts the least, and lifts presence the softest and lowest. Values from the processor — there are no editable numbers.
| Profile | High-pass | Low-mid cut | Presence boost |
|---|---|---|---|
| Warm | 70 Hz (lowest) | -1 dB @ 350 Hz (gentlest) | +2 dB @ 2.5 kHz (softest, lowest) |
| Male | 80 Hz | -2 dB @ 300 Hz | +3 dB @ 3 kHz |
| Female | 100 Hz | -2 dB @ 300 Hz | +3 dB @ 4 kHz |
| Bright (broadcast) | 80 Hz | -1.5 dB @ 250 Hz | +4 dB @ 5 kHz |
What 'warmth' means here (and what it doesn't)
An honest map of how this profile creates warmth — by preservation, not by adding bass.
| Mechanism | Does Warm do it? | Detail |
|---|---|---|
| Keep more low end | Yes | 70 Hz high-pass passes more body than the 80-100 Hz profiles |
| Cut less low-mid | Yes | Only -1 dB at 350 Hz, the lightest cut of any profile |
| Soften the top end | Indirect | Lowest, smallest presence boost (+2 dB @ 2.5 kHz) avoids adding harshness |
| Add a bass boost | No | There is no low-shelf or bass-boost band in any profile |
Tier limits
Pro feature. Per-file; single file per job.
| Plan | Max size | Max duration | Access |
|---|---|---|---|
| Free | 50 MB | 30 min | Locked |
| Pro | 200 MB | 120 min | Yes |
| Pro-media | 100 GB | Unlimited | Yes |
| Developer | 100 GB | Unlimited | Yes |
Cookbook
Warming thin voices in real situations, each with the exact FFmpeg chain.
Audiobook narration from a budget condenser
Warm keeps body (70 Hz HPF), barely touches the low-mid, and adds just a touch of presence — ideal for long-form listening that shouldn't fatigue the ear.
Profile: Warm highpass=f=70, equalizer=f=350:t=q:w=1:g=-1, equalizer=f=2500:t=q:w=1:g=2 chapter-01.wav -> chapter-01-eq.wav
Tinny phone-recorded voice memo
A phone mic is all treble. Warm's preserved low end and soft presence make it fuller without amplifying the harsh top.
Profile: Warm memo.m4a -> memo-eq.m4a (re-encoded AAC) Then, if still harsh: /audio-tools/de-esser
Warm sounds boomy — switch to Male
If the source already had body, Warm's preserved low end can tip into boom. Male's 80 Hz high-pass and -2 dB cut tighten it back up.
Tried: Warm -> boomy Use instead: Male highpass=f=80, equalizer=f=300:t=q:w=1:g=-2, equalizer=f=3000:t=q:w=1:g=3
Don't expect bass to be added
Warm preserves low end; it does not synthesize it. If the voice has no low end at all (e.g. an aggressively high-passed source), there's nothing for Warm to keep.
Warm = lowest HPF (70 Hz) + lightest cut. No low-shelf boost exists. If the source was already cut at 200 Hz, no profile here puts that body back.
Warm then level for a cosy narration master
EQ for warmth, then even the level and normalise for a smooth listen.
1. voice-eq (Warm) 2. /audio-tools/speech-leveler 3. /audio-tools/loudness-normalizer (-16 to -18 LUFS)
Edge cases and what actually happens
Expected a bass boost, got a subtle change
By designWarm has no low-shelf or bass-boost band. It sounds warmer because it keeps more low end (70 Hz high-pass) and cuts the least (-1 dB), not because it adds bass. If you need genuine added low end, this chain can't do it.
Warm now sounds boomy
Wrong profileIf the source already had plenty of low end, preserving even more tips it into boom. Switch to Male (80 Hz high-pass, -2 dB cut) to tighten it.
Voice is still harsh after Warm
De-ess afterWarm is the gentlest profile but can't undo a harsh mic. Run the de-esser afterwards to soften the harsh S/top end while keeping the warmth.
Source was already heavily high-passed
No body to recoverIf a previous step cut everything below ~200 Hz, there's no low end left for Warm to preserve. EQ can't restore frequencies that aren't in the file — re-export from an earlier master if you have one.
No manual EQ controls
By designThe only control is the four-profile picker. Each is fixed. There's no way to nudge the high-pass frequency or boost a band yourself; use a DAW for that.
Free user can't open it
Pro requiredGated at Pro despite 'online'/'free' in the URL. Free locks the tool and caps the audio family at 50 MB / 30 min.
Recording over the duration limit
400 — duration exceededPro caps at 120 minutes per file. A long audiobook chapter set may exceed it — split with the audio-splitter, warm each part, rejoin with the audio-merger.
MP3 in, slight re-encode loss
ExpectedEQ re-encodes to the same container. MP3 in becomes a re-encoded MP3 out. Warm a WAV/FLAC master if available.
Stereo file with a music bed
SupportedFilters apply across all channels, so a music bed gets warmed too. EQ the voice before the bed is mixed in if you want only the voice affected.
Very low fundamental male voice
SupportedA very deep voice may already be full; Warm could push it boomy. Compare against Male and pick whichever keeps the voice intelligible, not just full.
Frequently asked questions
How do I warm up a thin voice recording?
Use the Warm profile. It keeps more low end (70 Hz high-pass), cuts the least low-mid, and uses the softest presence boost, so the voice sounds fuller without getting harsher or boomy.
Does Warm add bass?
No. There's no bass-boost band. Warm sounds warmer because it removes and cuts the least, preserving the body the voice already has.
Why not just boost the bass?
A crude bass boost adds boom and one-note muddiness. Preserving the natural low end and trimming less, as Warm does, gives a cleaner, fuller result for speech.
What if Warm sounds boomy?
The source probably already had enough low end. Switch to Male, which uses an 80 Hz high-pass and a -2 dB low-mid cut to tighten things up.
Can I adjust the warmth amount?
No — it's a fixed curve. The only choice is which of the four profiles to apply.
What if it's still harsh?
Run the de-esser after EQ. Warm is gentle but can't fix a harsh microphone on its own.
Is my recording uploaded?
No. It's processed in your browser with FFmpeg WASM and never leaves your device.
Is it free?
Free of upload and cloud cost, but the tool requires a Pro plan. Free doesn't include it.
Which formats work?
MP3, WAV, FLAC, M4A, OGG, and Opus. The output keeps the input format.
Can Warm fix a phone-mic recording?
It helps by adding body and softening the boost, but a phone mic's fundamental thinness has limits. Pair with the de-esser for harsh S sounds.
Can I process several chapters at once?
No — one file per job. For a batch chain, use podcast-master.
How long a file can I warm up?
Up to 120 minutes on Pro; unlimited on Pro-media and Developer.
Privacy first
Every JAD Audio tool runs entirely in your browser via FFmpeg (WebAssembly) and RNNoise. Your audio files never leave your device — verified by zero outbound network requests during processing.