One episode → a YouTube-ready chapter list: timestamped "MM:SS Title" lines you can paste straight into your show notes or video description.
Files are encrypted
EXAMPLE
EP-142.MP3 — 58 MIN EPISODE
arrow_forward
EP-142-CHAPTERS.TXT
00:00 Introduction
03:15 Why the launch slipped
17:40 Pricing the new tier
ep-142-chapters.txt · 9 chapters chapters
What you get
check_circle6 to 15 chapters, each marking a real topic shift
check_circle"MM:SS Title" lines starting at 00:00, ready for YouTube
check_circleShort, concrete titles — eight words or fewer
check_circleVideo files handled — the audio is extracted first
check_circleTimestamps from minute-level speech chunks, not word-exact
How it works
1Upload one episodeDrop in an audio or video file up to 500 MB — a podcast episode, a video, or a webinar recording.
2We transcribe, then pick the chaptersThe speech is transcribed on our own servers in minute-level chunks, then a language model picks 6 to 15 chapters at the real topic shifts and titles each one.
3Download the chapter listA plain-text file named after your episode, with "MM:SS Title" lines starting at 00:00 — paste it into a YouTube description or your show notes.
Who uses this tool
PODCASTING
Show notes with timestamps
Turn an episode into a chapter list for the show notes, so listeners can jump to the part they came for.
YOUTUBE
Chapters without the scrubbing
Paste the list into your video description and YouTube renders it as chapters on the progress bar.
MARKETING
Webinars people can navigate
A recorded webinar becomes a chaptered replay, so prospects can skip straight to the demo or the pricing.
EDUCATION
Lectures, split by concept
Mark where each new concept starts so students can revisit one explanation without replaying the hour.
INTERVIEWS
One chapter per question
Long-form interviews get a marker at each question, so the best answers are easy to find and share.
MEDIA & NEWS
Episode rundowns, timestamped
A news or panel episode becomes a rundown of segments with the minute each story starts.
SECURITY
Built for sensitive documents
Bank statements, medical files, case records. Security isn't a feature we added — it's the foundation.
lock
Encrypted in transit
Encrypted in transit over TLS 1.2+, and stored in access-controlled, encrypted object storage. Files are protected the moment they leave your browser.
admin_panel_settings
Your files stay yours
Workspaces are isolated per account. Role-based access shows teammates only what they need.
auto_delete
Deleted, not stored
Files are deleted after processing. Everything runs on our own hardware and is never sent to an outside AI service, so your data is never used to train models.
dnsProcessed on our own hardwareauto_deleteDeleted after processinglockEncrypted in transit
Questions
Common audio and video recordings: MP3, M4A, WAV, AAC, FLAC, OGG, OPUS, WMA, and video like MP4, MOV, M4V, WEBM, MKV, and AVI. For video we extract the audio track first, then transcribe the speech. One episode per job, up to 500 MB.
One plain-text file with 6 to 15 chapter lines in the form "MM:SS Title" ("H:MM:SS" past an hour), in chronological order, with the first chapter at 00:00. That is the format YouTube reads from a video description to show chapters on the progress bar, and it works as-is in show notes too.
Chapter timing comes from minute-level speech chunks, not word-level alignment. Each chapter lands on the right minute of the episode — close enough for show notes and YouTube chapters, but if you need a marker on the exact word, nudge it in your editor before publishing.
Yes. Your instructions can shape the titles and focus — shorter titles, a chapter for the ad break, one chapter per question. They are additive only: the output stays a chronological chapter list starting at 00:00 no matter what you ask for.
6 credits per minute of audio. Because the length is not known until the file is processed, the upfront hold is estimated from file size at a typical bitrate for the format (about 1 MB per minute for MP3 audio, more for WAV, FLAC, or video). The minimum is 6 credits.
Yes. Recordings run on our own servers, are never used to train any model, and are auto-deleted 7 days after the job finishes.