Paid submissions are currently closed. Uploading and your library are open.
Building the sound library for audio AI

The sound of
the world,
in one library.

Contribute recordings of voices, nature, machines, music and everyday noise. Help build our sound library and, when funded collections open, earn for accepted recordings.

Upload a recording Follow a sound
Illustrative sound shapes by category. Tap one to hold the view.
3 / 8
Voices Conversations Nature Animals Vehicles Machines Music Everyday noise Voices Conversations Nature Animals Vehicles Machines Music Everyday noise
Illustrative workflow · future stages are not built

From noise to something a machine can learn.

Built Not built yet
A closer listen · browser-local analysis

Every sound leaves a fingerprint.

Explore how frequency energy changes over time. Try a generated demo chime or inspect a short audio file in this browser.

WAV, MP3, M4A, OGG, FLAC · up to 20 MiB and 60 seconds

Audio is analyzed in this browser. It is not uploaded, saved, or used for training.

Choose a sound or try the demo chime to begin.

Frequency × time No sound selected
Energy · dBFS mapped to color
Audio frequency energy over time The plot will show measured frequency energy from the selected sound. Choose audio or try the demo chime. A sound’s measured pattern will appear here
Duration—
Decoded rate—
Channels—
Strongest displayed bin—

48 sampled moments from a mono mix. This is not a quality or authenticity score.

0:00
For creators · available today

Drop it in. Tag it. Find it again.

Your recordings stay exactly as you uploaded them.

Step 1

Sign up & upload

Sign up and drop in any recording. The original is stored untouched.

Step 2

Add a transcript optional

Write what’s said, if anything is.

Step 3

Categories & tags

Label it once, find it forever.

Step 4

Search, play, download

Your library, whenever you need it.

Always

Private originals

Your recordings belong in your library — not on display.

Original, untouched Stored exactly as you uploaded it. Nothing is overwritten.
You + authorized operators No one else can open your files.
Yours to take back Search, play and download any time.
After you upload · live today

Every recording starts with a closer look.

Fliq checks basic audio quality and exact duplicates. Eligible speech submitted to a funded offer goes to our team for review.

Illustrative review queue
WaveFileCategoryQualityStatus
rain_tin_roof.wav Nature Stored
market_chatter_02.wav Conversations In review
rain_tin_roof (1).wav Nature Exact duplicate
metro_brakes.m4a Vehicles Low quality
koel_morning.wav Animals Stored
Paid submissions Closed Payment requires an eligible, funded offer and approval.
Where we're heading

The journey of a sound.

Today we collect and organize. Our goal: turn reviewed, permissioned recordings into datasets for our own audio models.

Built Not built yet
01 · Built

Original recording

Your upload — raw, real and never altered.

02 · Not built yet

Cleaned copy

A separate, cleaned version beside the original.

03 · Not built yet

Permissioned dataset

Reviewed recordings with clear rights.

04 · Not built yet

Our audio models

Models that learn from the library.

Also not built yet Automatic cleaning Payments for general sounds Reliable fake-audio detection Model training
FAQ

Straight answers.

What's live, what isn't, and what happens to your audio.

Paid submissions are currently closed. The pilot is for eligible speech only; an open, funded offer and approval are required. Other sound categories have no paid offer.

Press record
on your world.

Start a private library of your recordings today.

Create your account