Skip to main content
The transcript editor opens when you click a Completed transcript in Speech to text. The text sits in segments beside the player, follows the audio as it plays, and can be corrected in place. This page covers reading and playing, editing segments, find and replace, the keyboard, copying and links, adding pronunciations and sending text to other services.
A transcript open in the editor: segments with the spoken word highlighted, the Speakers panel on the right and the player at the bottomA transcript open in the editor: segments with the spoken word highlighted, the Speakers panel on the right and the player at the bottom

Read and play

The player at the bottom of the page plays the transcript’s audio or video. Its controls and keys are described in Lists and uploads.
  • Click a word to play from it. Click a segment’s time to play from the start of that segment.
  • Follow playback (on by default) keeps the word being spoken in view. When you scroll away while it plays, following pauses and Back to playing appears; click it, or press B, to return to the spoken word.
  • Text size sets the reading size: Small, Medium or Large. Your choice, like Follow playback, is remembered in this browser.
  • “Search this transcript” shows only the segments that contain what you type.
  • Show one speaker shows one speaker’s segments. See Speakers.
The Back to playing button over the bottom of the transcript, shown after scrolling away from the word being playedThe Back to playing button over the bottom of the transcript, shown after scrolling away from the word being played

Edit a segment

1

Open it

Click the pencil (Edit segment) on the segment, or move to it with the arrow keys and press E.
2

Change it

Correct the text. To give the segment to another speaker, pick one in the speaker list at the top of the form; that change saves as soon as you pick.
A segment open for editing, with the speaker list, Delete, Cancel and Save at the top and Add a segment after below the textA segment open for editing, with the speaker list, Delete, Cancel and Save at the top and Add a segment after below the text
3

Save

Click Save. Opened with E, Enter saves too (Shift+Enter starts a new line). Cancel or Escape leaves without saving.
In the same form:
  • Add a segment after opens an empty segment below this one. Type its text, pick its speaker and save. A new segment has no word timings; it sits between its neighbours in time.
  • Delete segment asks Delete segment? and removes the segment and its text.
A segment can hold up to 10,000 characters. Longer text can’t be saved: “This segment is too long to save”. When you correct a segment, the words keep their timings where the text still lines up, so playback keeps following them.

Find and replace

Press Ctrl+F (Cmd+F on Mac) on the transcript page, or click Find and replace.
1

Find

Type in “Find in transcript”. The bar shows where you are, for example “3 of 12”. Enter goes to the next match, Shift+Enter to the previous one.
The find and replace bar over a transcript, with listen found 4 times, tune in as the replacement, and Replace and Replace allThe find and replace bar over a transcript, with listen found 4 times, tune in as the replacement, and Replace and Replace all
2

Set the options

Match case and Whole word are off by default. Ignore harakat is on: a search without vowel marks also finds the word written with harakat (and tatweel).
3

Replace

Type in “Replace with”, then Replace changes the current match, or Replace all changes every one.
Replace all asks first: Replace all matches? says how many matches change and in how many segments. When it’s done, the message “Replaced 12 matches” has an Undo button for about 10 seconds.
The Replace all matches? dialog saying how many matches change and in how many segments, with Replace allThe Replace all matches? dialog saying how many matches change and in how many segments, with Replace all
Some matches are left as they are, and the dialog or message says why:
  • A match in a segment you have open for editing. Save or cancel the edit to include it.
  • A replacement that would leave a segment empty.
  • A replacement that would make a segment longer than 10,000 characters.
Escape closes the bar.

Work from the keyboard

Click a segment (or Tab to it) and use these keys: Space, K, J, L, the left and right arrows, M and the speed keys stay the player’s. Letter keys don’t fire while you type in a field, and they work on an Arabic keyboard layout too (by key position). Press ? to see every shortcut on the page; see Navigation.
  • The copy button on a segment copies its text.
  • A segment’s More actions menu has Copy with timestamp ([01:24] Speaker 1: text) and Copy link at 01:24, a link that opens the transcript at that moment.
  • The Copy menu in the header has Copy transcript with timestamps (one line per segment, with time and speaker) and Copy link at the current playback time.
  • Shift-click a segment to select it, then Shift-click another to select every segment between them. The bar above the list shows “7 segments selected”, with Copy selection, Send to and Clear selection.
Three segments selected with Shift-click, and the bar above them with 3 segments selected, Copy selection and Send toThree segments selected with Shift-click, and the bar above them with 3 segments selected, Copy selection and Send to

Add a pronunciation

When a voice reads a word from your transcript wrong, save how it should be read to a dictionary. Voices then read it that way wherever that dictionary is on.
1

Pick the word

Select a word or phrase in a segment and click Add pronunciation beside it. You can also use the segment’s More actions menu, press D, or search “Add pronunciation” in Cmd+K (Ctrl+K on Windows).
2

Say how it's read

Check Word or phrase and type Read as, how it should sound.
3

Pick the dictionary

Choose a Dictionary, or New dictionary and give it a name.
4

Save

Click Save. “Saved to …” names the dictionary. If the word is already in it, the dialog says so and saving replaces the old reading.
The Add pronunciation dialog with the phrase Behind the mic, Read as be-hind the mike, and the Brand names dictionary pickedThe Add pronunciation dialog with the phrase Behind the mic, Read as be-hind the mike, and the Brand names dictionary picked
The pronunciation doesn’t change the transcript’s text.

Send text to text to speech or a voice over

From the header’s Copy menu, a segment’s More actions menu or the selection bar, choose Send to text to speech or Use in a voice over. The text goes without times or names, each speaker’s turn as a paragraph.
A segment's More actions menu: Copy with timestamp, Copy link at 00:00, Send to text to speech, Use in a voice over and Add pronunciationA segment's More actions menu: Copy with timestamp, Copy link at 00:00, Send to text to speech, Use in a voice over and Add pronunciation
  • Text to speech takes up to 3,000 characters at a time and a voice over script up to 1,000. For longer text, Hamsa Media offers to use only the first characters up to the limit, or to send your selection instead.
  • If you were already writing a script there, Hamsa Media asks before replacing it.
See Generate speech and Create a voice over.

Rename the transcript

Click the title at the top of the page, type the new one (up to 150 characters) and press Enter. Escape cancels. The title can’t be empty.
The transcript title open for editing at the top of the page, with its length 51/150 belowThe transcript title open for editing at the top of the page, with its length 51/150 below

Good to know

  • Long transcripts load in pages as you scroll; find, export and copy fetch the whole transcript first.
  • Edits save to the transcript at once. There is no history or undo for a segment edit; only Replace all has an undo.
  • Every member of the project can edit transcripts, whatever their role. The permissions page has no switches for speech to text. See Roles and permissions.

Speakers

Rename, add, merge and filter speakers.

Export a transcript

Subtitles, documents and data.

Dictionaries

Where pronunciations live and apply.

Lists and uploads

The media player and its keys.