01
Cleanup you can’t hear
Filler words, room hum, mouth clicks, breath spikes. Cuts land at sentence boundaries, never mid-word, so the edit doesn't show. You hear the take.
Production · the boring, fixed
The work after “stop record”, done while you breathe
Mato cleans every episode, levels it to platform spec, writes the captions, and marks the chapters. Renders go out to every destination. You either publish or review.
Episode 47 · pipeline
Running · 62%
- Trimsilence + ums
- Denoiseroom + breath
- Level−16 LUFS
- Caption99 langs
- Render5 formats
- 00:01.4trimremoved 2,140 ms of silence (47 segments)
- 00:18.2denoise−18 dB room tone, breath gate at −42 dB
- 00:34.8levelintegrating loudness… target −16 LUFS
- Pipeline
- 62%
- Elapsed
- 1m 48s
- LUFS now
- −16.0
- Caption confidence
- 99.4%
- Spotify-ready loudness
- −16 LUFS
- Render per 60 min episode
- ~2 min
- MP3, WAV, MP4, vertical, audiogram
- 5 fmts
What gets done
Every tedious step, handled before you touch the file
02
Loudness to platform spec
Integrated loudness normalized to −16 LUFS for podcast platforms, with true-peak limiting at −1 dBTP. YouTube and TikTok renders re-targeted to their own specs automatically.
03
Transcripts & chapters
Word-accurate transcript in 99 languages. Chapter markers inferred from topic shifts and embedded in MP3 ID3, MP4, and YouTube descriptions. Editable before publish.
Loudness
Mastered to spec
Each platform has its own loudness target. Mato measures integrated LUFS, true-peak, and dynamic range, then normalizes once per destination. You upload once. Each render lands at the right level for where it’s going.
- −16 LUFS for podcast feedsSpotify, Apple Podcasts, iHeart, Pocket Casts
- −14 LUFS for streaming videoYouTube and Twitch render targets
- True-peak limited at −1 dBTPSo nothing clips after platform encoding
Mastering · ep 47
Compliant
- Integrated
- −16.0
- Short-term
- −17.2
- True peak
- −1.4
- Range (LRA)
- 7.8 LU
- Crest factor
- 14 dB
Trim · ep 47
Sentence boundaries only
Raw take
After the pass
2,140 ms of silence removed across 47 segments
Sentence-boundary cuts
Filler removed where it doesn’t show
Every cut sits at a sentence boundary. Crossfades match the room tone. “Um,” “uh,” long silences, false starts: gone. Coughs and stumbles too, if you want. Your cadence stays intact.
- Crossfades matched to room tone
- Sentence-boundary cuts only. No mid-word clipping
- Per-host sensitivity (some hosts say "um" on purpose)
Multi-format render
One source, five deliverables
From every finished episode Mato renders an MP3 for the feed, a WAV for archive, a horizontal MP4 for YouTube, a vertical 9:16 cut for Shorts, and an animated audiogram for socials. Each at the right specs for its destination.
- MP3 + WAVFeed-ready and archive · chapters embedded in ID3
- 16:9 MP4 with auto-captionsYouTube specs, with chapter timestamps in description
- 9:16 vertical clipFor Shorts, Reels, TikTok · How clips are picked
Pairs with
More of the boring stuff, gone
01Transcripts & chaptersSearchable, editable, embeddedWord-level transcript in 99 languages, chapter markers inferred from topic shifts.02Clip generationFive short cuts, auto-pickedMato finds the funniest, sharpest, and most quotable moments and renders them vertical.03DistributionOne click. 150+ destinationsAudio platforms, video, socials. Mato hands the right render to the right place.
Stop polishing.
Start publishing
Upload a raw take. Mato hands you a feed-ready episode with transcript, chapters, and clips, in the time it takes to make coffee.