Can the anchor be told apart from the guest?
Yes. With the split by speaker on, each voice gets a label of its own, so the questions of the anchor and the answers of the guest stay apart.
News broadcast transcript: the broadcast written out with timecodes, so the anchor, the guest and the report can be quoted and found. Free $0, 3 files a day.
Upload the recording of the broadcast and it comes back written out, with the time of every line and the anchor and the guest named apart. One file is up to 30 minutes; the free plan is $0 with 3 files a day.
A broadcast is written down twice: once while it is spoken, and once by the people who need it afterwards. The second version is where the trouble starts, because a reporter who quotes a sentence heard it in a clip of thirty seconds and heard it in a hurry. A transcript of the whole broadcast gives the exact wording and the moment it was said, which is the thing a fact-check needs.
| What | Value |
|---|---|
| One broadcast | up to 30 minutes on the free plan, no minute limit on Pro |
| The anchor, the guest and the report | each voice gets a label of its own when the split by speaker is on |
| The text of the ticker | not in the file: only the sound of the broadcast becomes text |
| The free plan | $0: 3 files a day, up to 30 minutes per file, no card |
The file the broadcaster or the archive gives you is the one to upload. A broadcast 30 minutes on the free plan has to be split.
Send the file to the widget above, set the language or leave it on automatic detection, and turn the split by speaker on.
Search the text for the phrase and read it with its timecode, so the quote can be checked against the moment it was said.
Export TXT for the quote, CSV when the segments have to be listed for an archive, and DOCX on Pro for the document.
Keep the file of the whole broadcast, not only the clip.
Note the timecode of every quote you are going to use.
Name the file after the programme and the day of the week.
Export CSV when the archive wants one row per segment.
Yes. With the split by speaker on, each voice gets a label of its own, so the questions of the anchor and the answers of the guest stay apart.
Neither is read. The service transcribes the sound of the file: the text of the ticker and the captions burned into the picture are not part of it.
up to 30 minutes, the same on every plan. The free plan allows 3 files a day and 30 minutes per file.
No. It transcribes what was said. The check of the claim against other sources is the job of the newsroom.
Speakers by name, a time on every line, export to TXT, SRT, VTT, CSV and Markdown.