Subtitles

Free Subtitle Generator: Audio and Video to SRT and VTT

Upload a video or audio file and SpeakScribe transcribes it with Whisper and turns the timestamps into subtitle files: SRT and WebVTT, ready for YouTube and video editors. You also get the plain and timestamped transcripts. Guests can try it without an account.

New uploads are paused until it finishes, so the service stays fast for everyone. You can leave this page — the result will be waiting in your queue.

Open my queue →

MP3, M4A, WAV, FLAC, OGG, MP4 · up to 20 MB as a guest

Transcribing your audio…

How it works

1

Upload the video or audio

MP4, MP3, M4A, WAV, FLAC, OGG or WMA. Choose the spoken language for the best result.

2

Download the subtitles

When the transcript is ready, download Subtitles (.srt) or Subtitles (.vtt).

3

Check and upload

Fix names and terms in any text editor, then upload the file to YouTube or import it into your editor.

SRT and WebVTT

SRT for YouTube, Premiere Pro, DaVinci Resolve and most players; WebVTT for web video players.

Readable cues

Long passages are split into cues of up to two lines (about 42 characters each), the length most subtitle guidelines use.

99 languages

Whisper recognises 99 languages — the subtitles come out in the language that was spoken.

Getting it right

SRT or VTT?

Both contain the same cues. SRT is the most widely supported: YouTube, Facebook, video editors and players such as VLC read it. WebVTT is the format of HTML5 video players on websites. If you are not sure, use SRT.

Upload to YouTube

In YouTube Studio open the video → Subtitles → choose the language → Add → Upload file → With timing, then select the .srt file. You can edit any line in YouTube's editor afterwards.

Import into an editor

Premiere Pro imports SRT through File → Import and places it as a captions track. In DaVinci Resolve, import the file into the Media Pool and drag it onto the timeline. To test quickly, drop the .srt onto a playing video in VLC.

How the timing works — and its limits

Timing comes from the speech segments Whisper detects, so cues start and end with the speech. A long segment is split into several cues and its time is shared out by text length, which can drift by a moment within long sentences. Languages written without spaces (Thai, Khmer, Chinese, Japanese) are split by length rather than at word boundaries. Skim the file once before publishing.

All limits in one table: YouTube SEO from your transcript.

Great for

YouTube videos
Short-form video
Course lessons
Interviews and podcasts

Frequently asked questions

Guests can transcribe up to 20 minutes in total with files up to 10 minutes each; a free account includes 120 minutes a month with files up to 30 minutes. Every transcription includes the SRT and VTT files.
The subtitle files are in the spoken language. To make a translated version, use Translate transcript after transcribing, then pair the translated text with the original timing — or upload the original subtitles to YouTube and add a translated track there.
No. You get separate subtitle files, which viewers can switch on or off. To burn them in, import the SRT into your video editor.
MP4 directly. Convert MOV, MKV or WEBM to MP4 first, or upload just the audio as MP3 or M4A.

Do more with your content

Transcription, 361 AI voices, and YouTube SEO — all in one free creator suite.

Explore all tools