Powered by OpenAI Whisper — 50+ languages supported

Audio to perfect subtitles. Instantly.

Drop any audio file, get frame-accurate SRT captions for Premiere Pro — or clean plain text. Free, private, and blazing fast.

Start transcribing ↓ See how it works
Word-level timestamps SRT for Premiere Pro 50+ languages 100% browser-based No files stored 4 caption styles MP3 · M4A · WAV · WebM Free to use Word-level timestamps SRT for Premiere Pro 50+ languages 100% browser-based No files stored 4 caption styles MP3 · M4A · WAV · WebM Free to use
What makes it great

Everything you need for perfect captions

Professional-grade transcription with the flexibility video editors actually need.

⚡

Word-level precision

Whisper returns per-word timestamps so your captions sync to the exact millisecond — no more manually adjusting timing.

Frame accurate
🌐

50+ languages

Automatic language detection or manually specify the language. Supports English, Spanish, French, Arabic, Japanese, and dozens more.

Auto-detect
🔒

Fully private

Everything runs in your browser. Your audio goes directly to OpenAI's API — no intermediate servers, no stored files, no tracking.

Zero storage
🎬

Premiere Pro ready

Export industry-standard SRT files that import perfectly into Premiere Pro, DaVinci Resolve, Final Cut Pro, and any NLE.

SRT standard
Simple as 1-2-3

From audio to captions in under a minute

No software to install. No account required. Just bring your OpenAI API key.

1

Add your API key

Paste your OpenAI API key — it stays in your browser, never touches our servers. Whisper costs ~$0.006 per minute of audio.

2

Drop your audio file

Upload any audio or video file up to 25MB. MP3, M4A, WAV, WebM, OGG — all supported. Pick your caption style and language.

3

Download your SRT

Whisper transcribes and we generate your SRT instantly. Copy or download and drop it straight into your video editor.

Caption styles

Four modes for every workflow

Choose the caption style that fits your video — from punchy single words to full sentences.

Mode 01

Word by word

One caption per word — perfect for that viral, high-energy caption style seen on reels and shorts.

1
00:00:01,200 --> 00:00:01,600
Hello
2
00:00:01,600 --> 00:00:02,100
world
Mode 02

Words grouped

Group 2–5 words per caption for a natural reading rhythm without breaking the sentence flow.

1
00:00:01,200 --> 00:00:02,400
Hello world this
2
00:00:02,400 --> 00:00:03,800
is a test
Mode 03

Char-limited

Wrap at a max character count — ideal for broadcast standards or platforms with caption width limits.

1
00:00:01,200 --> 00:00:03,500
Hello world this is
a longer sentence.
Mode 04

Full segment

One caption per Whisper segment — great for narrative content and interviews where full sentences read better.

1
00:00:01,200 --> 00:00:05,800
Hello world, this is a longer
sentence for context.
FAQ

Common questions

Yes — this tool is completely free to use. The only cost is the OpenAI Whisper API fee, which is approximately $0.006 per minute of audio (~$0.36 per hour). For most videos this is just a few cents. You use your own API key so you pay OpenAI directly.
Yes. Your audio file is sent directly from your browser to OpenAI's servers — it never touches our infrastructure. We have no servers, no database, and no way to see your files or API key. Everything happens in your browser session only.
Whisper accepts MP3, MP4, M4A, WAV, OGG, WebM, and FLAC files up to 25MB. For larger files, compress or trim the audio first using a tool like Audacity or HandBrake. You can also extract just the audio track from a video to reduce file size.
In Premiere Pro, go to File → Import and select your .srt file. It appears as a caption track in your Project panel. Drag it onto your sequence timeline above the video track. You can then style the captions using the Essential Graphics panel. Works the same way in DaVinci Resolve and Final Cut Pro.
Sign up or log in at platform.openai.com, go to API Keys, and create a new secret key. You'll need to add a small credit balance (minimum $5) to your account. The Whisper API is extremely affordable — $5 will transcribe roughly 14 hours of audio.
Whisper large-v2 (used by the API) achieves near-human accuracy on clear audio — typically 95%+ word error rate on English. Accuracy improves significantly when you specify the source language manually rather than relying on auto-detection. Background noise, strong accents, or multiple overlapping speakers may reduce accuracy.

Ready to caption your audio?

Free to use · No sign-up · Instant results · Premiere Pro compatible

Launch the tool ↓
The tool

WhisperSRT Transcriber

Your audio never leaves your browser session.

1 OpenAI API Key
Your key is used only in this browser — sent only to OpenAI, never stored anywhere
2 Audio File
🎵

Drop your audio file here, or click to browse

MP3 · M4A · WAV · OGG · WebM · FLAC · up to 25MB

3 Caption Settings
Preparing... 0%
Cost estimate: Whisper charges ~$0.006/min. A 10-minute audio file costs about $0.06. Your key is stored only in memory for this session — it's gone when you close the tab.