Put music under a voiceover, podcast intro, presentation narration or audiobook clip. The music plays on its own for the intro, automatically dips (ducks) whenever you speak, swells back in the pauses, and fades out after the outro. The finished mix is loudness-normalized and saved as an MP3, all in your browser.
Last updated:
Showing a demo voice and demo music.
0:00 / 0:00
VoiceMusicMusic level (ducking & fades)
Advertisement
Advertisement
How to use the Add Background Music to Voice
Add your voice: drop a voice recording (MP3, WAV, M4A, video…) in box 1.
Add the music: drop a background track in box 2. Use music you have the rights to use. A demo voice and music are loaded so you can try first.
Set the music volume and ducking: −12 dB music with −12 dB ducking suits most voiceovers; raise Speech sensitivity if quiet words don't duck the music.
Choose intro and outro lengths and the fades, then press play and click the timeline to check any part.
Pick the loudness target and format and press Download mix.
Recommended levels for common uses
Levels are relative: the music volume sets how loud the bed plays in the intro and pauses, and ducking subtracts from that while you speak. Start here and adjust by ear.
Use
Music volume
Duck by
Attack / release
Loudness target
Podcast intro / outro
−8 to −10 dB
−14 dB
250 ms / 1.0 s
−16 LUFS (stereo) or −19 LUFS (mono)
YouTube voiceover
−12 dB
−12 dB
200 ms / 0.9 s
−14 LUFS
Tutorial or e-learning narration
−18 dB
−10 dB
200 ms / 1.2 s
−16 LUFS
Meditation or storytelling
−14 dB
−6 dB
400 ms / 2.0 s
−19 LUFS
Presentation or slideshow audio
−15 dB
−12 dB
250 ms / 1.0 s
−16 LUFS
Broadcast (EBU R128)
−12 dB
−12 dB
250 ms / 1.0 s
−23 LUFS
How the mix is built
Timeline: total length = intro + voice + outro. The voice starts after the intro and is placed in the centre of the stereo image.
Speech detection: the voice is measured every 10 ms. Frames louder than a threshold set by Speech sensitivity (12 to 42 dB below the voice's typical level) count as speech, held for 300 ms to bridge short gaps, and the ducking starts one attack time early (look-ahead).
Music level: music gain = music volume × ducking curve × fade curve. Fades use a quarter-sine shape, which sounds smoother than a straight line.
Loudness: the integrated loudness of the mix is measured (K-weighted, 400 ms blocks, −70 LUFS absolute and −10 LU relative gates) and a single gain brings it to the target.
Limiter: a 5 ms look-ahead limiter keeps sample peaks at or below −1 dBFS.
Frequently asked questions
What is audio ducking?
Ducking automatically turns the background music down while someone speaks and back up when they pause, the way radio presenters and podcasts keep speech clear without the music disappearing. In a studio it is done with a sidechain compressor. Here the voice track is analysed in 10 ms steps; when speech is detected, the music level glides down by the Duck by amount with the attack time, and recovers with the release time.
What settings should I use?
For a typical voiceover, music at −12 dB with −12 dB of ducking leaves the music clearly audible in the intro and pauses but well below the voice. For dense speech such as a tutorial, try −18 dB music and −10 dB ducking. Use a short attack (150–300 ms) so the first word is not covered, and a release of about one second so the music does not pump up between words.
What does loudness normalize do?
It measures the integrated loudness of the whole mix in LUFS (loudness units relative to full scale) using the ITU-R BS.1770 K-weighting and gating method, then raises or lowers the level to your target. Streaming platforms play audio at roughly −14 LUFS, many podcast guidelines use −16 LUFS for stereo and −19 LUFS for mono, and European broadcast uses −23 LUFS. A limiter keeps the peaks below −1 dBFS.
What if the music is shorter than my recording?
With Loop music ticked, the track repeats as long as needed, with a short 40 ms crossfade at the loop point to avoid a click. For the smoothest result, use music that was made to loop, or a track longer than your voice plus the intro and outro. You can also skip the start of the music with Start music from.
Can I use any song as background music?
Only use music you have the right to use, such as your own music, royalty-free tracks, or music licensed for your project. Commercial songs are protected by copyright, and platforms like YouTube may block or claim videos that use them. This tool does not provide music; it mixes the files you choose.
Is my audio uploaded anywhere?
No. Both files are decoded, analysed and mixed in your browser with the Web Audio API. The only download is the MP3 encoder library (lamejs), fetched once from a public CDN when you save an MP3; WAV export works completely offline.
Are all the tools on FreeToolCabin tested for quality and accuracy?
Yes! Every tool on our site undergoes automated Q/A testing in real browser environments to ensure high precision, correct output, and zero errors. If you happen to spot a bug or have a suggestion, please contact us at [email protected] and our team will review it right away.