Recording
Your recording
Paste your recording. Nothing to upload.
←recording
Make a transcript
Turn a recording into a transcript
Every segment timed, so any sentence can be found and played back.
Free to try, no card. Your recording opens in Ancher.
Transcript with timingsCommitment languageUnresolved threads
Kept in parts
Every one of these stays separately addressable instead of collapsing into one block of text.
Full textTimestampsFiller removedExport formats
Transcript
The spoken words, cleaned up, with a timestamp on every segment.
A transcript is only useful if it is anchored. Ancher transcribes the audio and puts a start and end time on every segment, cleans filler without paraphrasing, samples frames from anything shared on screen so a number that was pointed at survives, and states plainly that there are no speaker labels rather than inventing them.
The honest version
An unanchored transcript is a wall of text nobody can check
Without timings, a forty-page transcript of a ninety-minute call is unusable in a dispute — finding the sentence means rereading or rescrubbing. With a timing on every segment, a contested line becomes a click. That is the difference between a transcript that gets referenced and one that gets generated and never opened again.
One second of one recording, at 18:47, read two ways.Illustrative. The figure is invented; the gap it falls into is not.
if you look at the number on the right there, that's the one that changed —
On the shared screen — not readThe specific never reaches the draft
Transcript
On the shared screen
Q4 forecast: 1.9M
from 18:47
Lands in the draft, with its timecode
- Start and end seconds on every segment, not a timestamp every few minutes.
- Filler is removed without paraphrasing, so nothing becomes a summary of what was said.
- No speaker labels are invented; the transcript says so at the top.
What comes out
Transcript, section by section
The spoken words, cleaned up, with a timestamp on every segment.
Full text
The spoken words, cleaned of filler, with screen content noted where something was shared.
Timestamps
A start and end time on every segment, not a marker every few minutes.
Filler removed
What was removed, so nothing reads as a paraphrase of what someone said.
Export formats
Plain text you can paste into a document or a ticket.
Searchable in your workspace
Indexed with your other sources, so a search reaches this call by something said in it.
Recording-to-transcript prompt
The link on its own gets a summary. This is the instruction that produces the 5 sections above, in that order, with the rules that keep them honest. Paste it with your recording — in Ancher, or in whatever assistant you already use.
Transcribe this recording. (1) Give the spoken words cleaned of filler, without paraphrasing. (2) Put a start and end time on every segment. (3) Note what you removed in cleanup. (4) Where something was shared on screen, describe what was shown and when. (5) State at the top that there are no speaker labels and do not attribute any line to a person.
The detail, if you want it
How each part of a recording becomes a section
| From the recording | Into | Why |
|---|---|---|
| Transcript with timings | Timestamps | Every segment carries start and end seconds, which is what turns a claim into a record. |
| Transcript with timings | Full text | The spoken words are the document, and cleaning filler without paraphrasing is what keeps it a transcript rather than a summary. |
| Screen shares, if it is video | Full text | A shared screen is sampled as frames, so a figure that was pointed at rather than read out still appears. |
| Duration and timeline | Searchable in your workspace | Knowing the length lets a search result say where in a ninety-minute call the match is. |
| Transcript with timings | Filler removed | Removing filler is a decision, and recording what was removed is what keeps the transcript defensible. |
Everything Ancher keeps from a recording
Transcript with timings
Every segment carries start and end seconds, which is what turns a claim into a record.
Commitment language
I'll take that, by Friday, let's park it — these phrases are what separate a task list from a topic list.
Unresolved threads
What a conversation failed to close is usually the reason for the next one.
Screen shares, if it is video
A shared screen is sampled as frames, so the number that drove a decision stays attached to it.
Duration and timeline
Where the time actually went is a finding in itself for any recurring meeting.
The four steps
- Add the recording. Upload the audio or video file. Ancher transcribes it and keeps a start and end time on every segment, so any line in the output can be traced back to the moment it was said.
- Nothing gets flattened. Transcript with timings, commitment language, unresolved threads, screen shares, if it is video, duration and timeline stay separately addressable instead of collapsing into one block of text.
- Ask for a transcript. Ancher fills full text, timestamps, filler removed, export formats, searchable in your workspace using the mapping above, not a generic summary pass.
- Check it before you use it. Every specific keeps a link back to where it came from in the recording, so a claim can be verified without reopening the source.
Where this goes wrong
- There are no speaker labels. A four-person call reads as one continuous voice, and attributing a line to someone is your judgement rather than the transcript's.
- Crosstalk and room mics degrade accuracy on exactly the passages that matter most, which are the ones where people interrupt each other.
- Recording others carries consent and retention obligations that vary by jurisdiction, and a transcript is a more durable record than the audio.
Questions people ask
Does it separate speakers?
No. It gives you what was said and when. For anything that turns on attribution — a disciplinary record, a legal matter — this is the wrong tool and the transcript says so at the top rather than in a footnote.
How accurate is it?
Good on clear single-speaker audio and worse with crosstalk, room mics or accents the model handles poorly. The failure mode is worst precisely where people interrupt each other, which is where decisions tend to happen.
Why sample the screen?
Because in a screen-shared call the number is on the screen and the speech is that one right there. A transcript without the frames records a sentence that points at nothing.
Bring your own recordings
Everything you save lives in one workspace, so the transcript is built from your sources — not from a model's memory of the internet.
Open Ancher →Not a recording? The same transcript also comes from YouTube video, Instagram post, TikTok video.