Free AI subtitle generator

Free audio to SRT converter

Turn any audio recording into a timed SRT. Drop a file, get subtitle cues with short readable lines

3 AI models

4 free export formats

99 languages

Transcribe audio and video in 99 languages

  • English
  • Español
  • 中文
  • Français
  • Deutsch
  • 日本語
  • Русский
  • Português
  • Italiano
  • 한국어
  • العربية
  • हिन्दी
  • Türkçe
  • Polski
  • Nederlands
  • Български
  • বাংলা
  • Čeština
  • Dansk
  • Ελληνικά
  • فارسی
  • Suomi
  • עברית
  • Magyar
  • Bahasa Indonesia
  • മലയാളം
  • Română
  • Svenska
  • Kiswahili
  • தமிழ்
  • తెలుగు
  • ไทย
  • Українська
  • اردو
  • Tiếng Việt
How it works

audio to srt in 3 steps

  1. 1

    Upload your file

    Drop your file or click to choose. MP3, M4A, WAV, FLAC, OGG, and more.

  2. 2

    Pick language and model

    Auto-detect the language or choose from 99. Use free Turbo for speed, or Studio for the best accuracy.

  3. 3

    Export your subtitles

    Read the transcript in seconds, then export timestamped SRT subtitles ready for any editor.

Why Typist

Built for fast, accurate transcripts

An hour in about a minute

Groq-served Turbo runs at roughly 200x real time, so your transcript is ready almost immediately

Every export, free

Download as plain text, Word, PDF, or timestamped SRT subtitles on every plan

Your file stays yours

Uploaded only to transcribe, removed afterward, and never sold, shared, or used to train models

99 languages

Auto-detected or pick your own, with the most accurate model recommended per language

Beyond transcription

Your transcript is just the start

  • AI summary and key moments

    One tap turns the transcript into a TL;DR, key quotes, and action items.

  • Auto chapters

    Long recordings are split into navigable chapters you can jump between.

  • Share or export anywhere

    Send a clean public link, or export to TXT, DOCX, PDF, or SRT.

Summary
Chapters
IntroKey pointsQ&AWrap-up

Ready to turn audio into subtitles?

Drop a file and read your transcript in seconds. Free to start, no signup.

Transcribe a file
The format

What audio actually is

An SRT is a plain-text subtitle file: numbered cues with a start and end timecode and the caption text. Audio files carry no timing for captions, so Typist transcribes the recording with word-level timing and writes the cues, turning any audio into subtitles you can use over video or as standalone captions.

Audio is a family of containers and codecs (MP3, AAC, Opus, PCM), and Typist reads the common ones, so you never convert first. Lossy or lossless makes no difference for clear speech. The SRT quality tracks the recording, and cue boundaries come from word-level timing, then the words are grouped into short lines.

Where these files come from

Phone recordings, voice notes, podcasts, lectures, and interviews. Whatever a microphone captured that you want as timed captions.

  • Podcasts
  • Voice notes
  • Lectures
  • Interviews
How audio becomes textAccuracy comes from the audio, not the file type
  1. audio fileYour upload
  2. Audio decodedThe speech is what we transcribe
  3. TranscriptCopy or export to TXT, DOCX, PDF, SRT
Output
SRT subtitles
Timing
Word-level
Lines
~42 chars
Works with
Any editor
Subtitles

Timed captions, ready for your editor

why some projects stall
captions.srt
100:00:00,000 --> 00:00:03,200Today we are looking at
200:00:03,200 --> 00:00:06,400why some projects stall
300:00:06,400 --> 00:00:09,800even when the team works hard

Loads into your tools

  • CapCut
  • Premiere Pro
  • DaVinci Resolve
  • YouTube Studio
  • Final Cut Pro
  • VLC

Readable on screen

Typist re-segments long speech into short timed lines of about 42 characters, at most two lines per cue, so captions stay readable. It will not stuff a whole paragraph into one cue.

FAQ

Questions about converting to text

More

Other ways to transcribe

Convert to text

Generate subtitles

By language