Skip to content
FilmScribe
Audio and speech

What is speaker identification?

Speaker identification is separating, in a transcript, who said each line. In a recording with several voices, every passage gets a label, like Speaker 1 and Speaker 2, which later becomes each person's name. It is what turns a wall of text into a conversation you can actually read.

Also called: speaker labels, speaker separation

See transcription Try it with my video
Example

From labels to names

First the transcript separates the voices. Then the labels become the names of the people in the conversation.

Officiant 00:18:40 Ryan, go ahead and share the words you wrote for Emma.
Ryan 00:18:52 Emma, I promise to choose you again on every ordinary day.
Emma 00:19:30 You were the answer to a prayer I didn't know I was praying.
Why it matters

Why separate the speakers

In a podcast, an interview or a wedding, the question “who said that?” comes before any editing. With speakers separated, you filter to the guest, quote the right person and build the paper edit by character.

In vertical video, knowing who is speaking also drives the framing: the frame goes to whoever has the floor.

In FilmScribe

Speakers in FilmScribe

We separate the lines by person automatically, on every plan, and you filter the transcript by speaker. On Studio and Agency, you can rename each speaker and ask us to identify the names: we find the moment each person introduces themselves or is called by name and show that passage as proof, so you can check and apply it.

Transcription and speakers Every line with its speaker and timestamp, a filter by person and, on Studio and Agency, names identified with proof. See transcription →
In practice

Getting speakers labeled correctly

  1. 01

    Collect names in writing

    Ask for every participant's name, spelling and title in writing before the recording. A label is only as good as the name behind it, and a misspelled guest in a published quote is hard to undo.

  2. 02

    Have people say their names

    At the start of an interview, ask each person to say and spell their name on camera. It doubles as a slate and gives every label a clear name to match.

  3. 03

    Check short interjections

    Quick replies like “yeah” or “right” are easy to assign to the wrong person. Check them wherever they matter, like in a quote, and move or drop them.

  4. 04

    Confirm similar voices

    Two people with similar voices are easy to swap. Listen to a couple of lines from each label before you rename it, since a swap at the start carries through the whole transcript.

Related terms

See all terms →
Transcription The speech in audio or video written down as text, either verbatim, with every hesitation, or clean and easy to read. Timestamps Markers of the moment each line or scene happens in a recording, in hours, minutes and seconds. Reframe Cropping a horizontal video to vertical or square while keeping the person who is speaking in frame. Paper edit The edit planned on paper from transcripts and timecodes, picking and ordering lines before opening the timeline.

Frequently asked questions

Does the transcript know each person's name?

It separates the voices but does not know names on its own. On Studio and Agency, we read the transcript, find where each person introduces themselves or is addressed and suggest the name with the passage that proves it.

Does it work with lots of people talking?

It works with several voices. The more people talk over each other, the harder it is to separate them, so recordings with one microphone per person give the best results.

Can I fix a wrong speaker?

Yes. You rename speakers and edit the transcript text, and the fix carries over to your exports.

Captions, clips and paper edits. All in one place.

Upload a video or paste a YouTube link. We transcribe every word with its timing, and you take the result to social media or to your editor.

Try it with my video
See the result on your own recording before adding a card.