# Transcript-based editing

Canonical URL: https://getvidiyo.app/glossary/transcript-based-editing
Last updated: 2026-10-08

> Definition: Transcript-based editing is a way of editing video through its transcript: the software transcribes the speech with word-level timestamps, and deleting or rearranging text cuts or moves the matching audio and video.

Transcript-based editing turns a video into a document. You see what was said as text, and editing the text edits the video: strike out a rambling sentence and that footage is removed; drag a paragraph up and the clip moves with it. For speech-driven content like podcasts, interviews and talking-head videos, it's often far faster than scrubbing a timeline.

## How does transcript-based editing work?

Transcript-based editing works by linking every word to a moment in the media. The steps:

1. **Transcribe.** A speech recognition model (often based on [Whisper](https://getvidiyo.app/glossary/whisper-transcription)) converts the audio to text.
2. **Align.** Each word gets a start and end timestamp.
3. **Edit the text.** Deleting words creates cuts at those timestamps; moving text reorders clips.
4. **Render.** The edit is applied to the audio and video together.

Because the timing is word-level, the same data also drives [word-level captions](https://getvidiyo.app/glossary/word-level-captions) and features like filler word removal.

## How is transcript-based editing used for YouTube and short-form videos?

Transcript-based editing is mainly used for the first, subtractive pass on spoken content: removing false starts, picking between repeated takes, cutting tangents and finding short clips inside a long recording. Searching the transcript ("where did I mention pricing?") is often the fastest way to find a moment in hours of footage.

## Transcript-based vs frame-aware editing: what's the difference?

Transcript-based editing decides cuts from words alone; frame-aware editing also looks at the picture.

| | Transcript-only | Frame-aware |
|---|---|---|
| Knows what was said | Yes | Yes |
| Knows what's on screen | No | Yes |
| Catches a blink or glance at a cut | No | Can |
| Places captions clear of faces | No | Can, from measured face positions |
| Good for | Podcasts, first cuts | Talking heads with graphics, product demos |

## How does Vidiyo use transcripts?

Vidiyo transcribes your footage on your Mac and uses the word-timed transcript under the hood, but you don't edit a document: you ask in chat. The assistant reads the transcript as sentences marked kept or cut, picks word ranges to delete or tighten, and then checks the composed frames, faces and captions before reporting back. The edits land on a real timeline with layers you can adjust by hand. See [transcript-based vs frame-aware AI video editing](https://getvidiyo.app/blog/transcript-based-vs-frame-aware-ai-video-editing) for a longer comparison.

## Frequently asked questions

### How accurate is transcript-based editing?

It's as accurate as the word timings. Good speech recognition gets most words right, but fast speech, crosstalk and music can shift a cut by a few frames. Most tools let you nudge cuts on a timeline afterwards.

### What can't transcript-based editing do?

It only knows what was said. It can't see whether you blinked at the cut, looked away, held up the wrong product, or whether a caption covers your face. Visual decisions need the frames.

### Is Vidiyo a transcript-based editor?

Partly. Vidiyo uses a local, word-timed transcript to find and cut words, but you edit by chatting rather than editing a document, and the assistant also looks at the actual frames before and after it cuts.

## Related terms

- [Whisper transcription](https://getvidiyo.app/glossary/whisper-transcription)
- [Agentic video editing](https://getvidiyo.app/glossary/agentic-video-editing)
- [AI video editor](https://getvidiyo.app/glossary/ai-video-editor)
- [Word-level captions](https://getvidiyo.app/glossary/word-level-captions)

## Where Vidiyo does this

- [Edit video by chatting](https://getvidiyo.app/features/chat-video-editing)
- [Remove silences and filler words automatically](https://getvidiyo.app/features/silence-and-filler-word-removal)
- [Transcript-based vs frame-aware AI video editing](https://getvidiyo.app/blog/transcript-based-vs-frame-aware-ai-video-editing)