Script-first audio and captions

Text to speech with SRT or WebVTT subtitles

Create an MP3 voice track and an SRT or WebVTT subtitle file from the same script in one generation. Download both files, then review their alignment in your video editor before publishing.

  • MP3 plus subtitle file
  • SRT or WebVTT output
  • One script, one generation

Listen to the fixed English sample, or generate and download up to 500 Unicode characters without an account. Guests get two attempts per shared network each day, including failed attempts; the limit resets at 00:00 UTC. Signed-in accounts keep the existing one-time 40 Credits for 30 days; each started 100 characters costs 1 Credit.

English voice sample
1

Enter text

2

Voice and delivery

1.0×
.5×2×
100%
0%200%
0Hz
-50+50

Guest text is limited to 500 Unicode characters per try and two attempts per shared network each day, including failed attempts. It resets at 00:00 UTC. Signed-in accounts keep the existing one-time 40 Credits for 30 days and 1 Credit per started 100 characters.

Real product sample

Listen to a generated MP3 and inspect its captions

This 14.4-second MP3 was generated from the script below. Its original SRT contains 37 word-level cues. The WebVTT download is a format conversion of that SRT with the timing preserved. A separate four-cue file groups the original script into sentences for editing.

Demo scriptDownload text
Turn a short script into a clear product demo. First, choose a voice and check the pronunciation. Then download the MP3 and subtitle file. Import both into your video editor, and review the timing before you publish.
First 4 of 37 SRT cuesDownload full SRT
1
00:00:00,075 --> 00:00:00,275
Turn

2
00:00:00,275 --> 00:00:00,300
a

3
00:00:00,300 --> 00:00:00,650
short

4
00:00:00,650 --> 00:00:01,038
script
First 4 WebVTT cuesDownload full WebVTT
WEBVTT

1
00:00:00.075 --> 00:00:00.275
Turn

2
00:00:00.275 --> 00:00:00.300
a

3
00:00:00.300 --> 00:00:00.650
short

4
00:00:00.650 --> 00:00:01.038
script
Four-sentence editing exampleDownload 4-cue SRT

This four-cue SRT is used as the sentence-grouped display caption file in the edited workflow demo. It is an editor-created example rather than an original TTSMint output, and it has not been tested as a CapCut import. The original CapCut technical-check MP4 was exported with the original 37-cue SRT; the separate edited demo appears in our CapCut import tutorial.

These word-level cues are useful for checking alignment, but the single-word lines are not ready-to-publish captions. In your editor, combine words into readable lines and review punctuation, names, pronunciation, and timing before publishing.

Generate voice and captions in three steps

The subtitle option is already enabled on this page.

01

Add and review your script

Use punctuation and short sentences to make pauses and caption breaks easier to review.

02

Choose SRT or WebVTT

Use SRT for many editors and players, or WebVTT for a web-video workflow. Choose your voice and delivery settings too.

03

Download and check the pair

Generate once. In the result panel, play the MP3 and download it together with the selected subtitle file. Short text can be generated without signing in.

Script-first TTS

Your submitted script is the source for both the new voice track and its subtitle cues. This avoids running a second transcription pass over the generated audio.

  • Useful for narrated videos and lessons
  • Separate MP3 and subtitle downloads
  • Same text source for audio and captions

Not speech-to-text

TTSMint does not accept an existing recording on this page. Use a transcription product when you need captions for audio or video that has already been recorded.

  • Review timing and line breaks before publishing
  • Check names and pronunciation
  • Guest generation allows up to 500 Unicode characters per try and two attempts per shared network each day, including failed attempts; it resets at 00:00 UTC. Signed-in accounts continue to use Credits.

Check the audio and captions before handoff

Open the MP3 and its matching SRT or WebVTT file in your video editor. Listen from the beginning and check a cue near the middle and another near the end. Correct line breaks, names, pronunciation, and cue timing in the editor if needed. For word-level output, merge adjacent cues into readable phrases or sentences and restore punctuation; an editable four-sentence SRT example is available with the sample files above.

  • Keep the MP3 and subtitle file from the same generation together.
  • Choose SRT or WebVTT based on the import formats listed by your target editor or player.
  • After editing the script and generating again, replace both files with the new pair.
  • If the gap grows from beginning to end, check timing at several points; shifting the whole subtitle track only addresses a consistent offset.

Speech and subtitle questions

What this workflow creates, which format to choose, and what to review.

How do I generate speech and subtitles together?

Enter your script, keep Generate subtitles enabled, choose SRT or WebVTT, and generate. After a successful request, the result panel provides separate MP3 and subtitle downloads.

What is the difference between SRT and WebVTT?

Both formats store timed caption cues. Choose SRT when your video editor or player lists SRT as an import format; it is widely supported by editors and players. Choose WebVTT for a web-video workflow when the target player or platform asks for .vtt. Check the import menu before generating or converting files because support varies by editor.

Are the subtitles created from the same script as the audio?

Yes. TTSMint creates the speech and subtitle timing from the submitted script in the same generation request. It does not transcribe an existing recording.

Do I need to review the subtitle timing?

Yes. Preview the MP3 with its subtitle file and check the beginning, middle, and end. If word-level cues are too short to read, merge adjacent cues into phrases or sentences and restore punctuation; keep the first cue start and last cue end for each group. The four-sentence sample above is an editor-created example, not the original generated SRT.

What if subtitles are offset or drift out of sync?

First confirm that the MP3 and subtitle file came from the same generation. Compare a cue at the beginning, middle, and end: a similar offset throughout may be corrected by shifting the subtitle track, while a growing offset points to changing timing and needs cue-by-cue or section edits. If you change the script, regenerate and replace both files as a pair.

Can I add subtitles to an audio file I already have?

No. This page is a script-first text-to-speech workflow, not speech-to-text. It creates new audio and captions from your text rather than transcribing an uploaded recording.