How it works

From file to subtitle in four steps

📁

Upload

Drop in any audio or video — MP3, WAV, MP4, MKV, MOV, WEBM, and more.

🧠

Pick a model

Download once, use forever. Choose Tiny through Large based on your hardware.

⚡

Transcribe

Runs on your CPU. Auto-detects language. Edit any segment directly in the app.

📤

Export

Export in 5 formats, translate into 20 languages offline, or burn subtitles into video.

Features

Everything you need to
work with spoken content.

A focused toolset for podcasters, researchers, filmmakers, journalists, and educators.

🎙️

Accurate Transcription

Industry-leading Whisper AI transcribes with professional accuracy across 60+ languages. Voice Activity Detection filters silence for cleaner results.

VAD Filter Word Timestamps Auto Language Detect

✏️

Editable Segments

Every transcribed segment is editable. Fix words, review timecodes, and search across the full transcript — all within the built-in editor.

Inline Edit Timecode View Full-text Search

🌐

Offline Translation

Translate subtitles into 20 languages with a translation model that runs on your own computer. No account, no API key, and nothing uploaded. Progress-tracked and resumable.

20 Languages Runs Locally No API Key

🎬

Burn Subtitles into Video

Permanently embed subtitles into MP4. Customize font, size, color, and position. Dual-subtitle mode shows original and translation simultaneously.

Dual Subtitles Font & Color Top / Mid / Bottom

🗣️

Voiceovers

Generate spoken audio for a translated subtitle track, preview each line, and regenerate any line on its own. Timing is fitted to the original subtitle window automatically.

Per-line Preview Auto-fit Timing Many Voices

🔒

100% Private

Transcription runs entirely on your CPU. Your files never leave your machine. No account. No data collection.

No Cloud Upload No Account Private by Design

Privacy by design

Your files stay
on your machine.

Transcription and translation both run on your own machine. The only time TranscribeKit uses the internet is when you choose to download a speech or translation model. There is no account, no API key and no upload.

Files uploaded to cloud

60+

Languages supported

Export formats

Translation languages

App screenshots

See the local transcription workflow

Add a video, download a Whisper model, edit timestamped transcripts, translate captions offline, export subtitles, and burn them into video from one Windows app.

TranscribeKit video library showing transcription status for each file — Video library

Every video you add to TranscribeKit, showing transcription status at a glance.

Editable transcript editor with timestamped subtitle segments — Transcript editor

Edit subtitle text against timecoded segments while the video plays alongside.

Side by side subtitle translation into Chinese running offline — Offline translation

Translate captions side by side into 20 languages, entirely on your own computer.

Subtitle export formats and burn-in settings with a rendered preview frame — Export and burn-in

Export SRT, VTT, TXT, JSON and TSV, or burn captions into the video with a live preview.

Whisper speech model management with download progress — Speech models

Download the Whisper model that suits your machine once and keep it on disk.

Offline translation model management and engine status in settings — Translation engine

Pick a translation model size, watch the engine status, and free memory when done.

Whisper models

Pick your precision

Download once, stored locally. Choose based on your hardware and how fast you need results.

Tiny

~39 MB · Fastest

Ultra-fast, minimal resource usage. Great for quick drafts or low-RAM devices.

Base

~141 MB · Very fast

Good balance of speed and accuracy. Ideal for everyday podcasts and meetings.

Small

~461 MB · Fast

High accuracy for most use cases. Recommended starting point for most users.

Medium

~1.5 GB · Moderate

Near-professional accuracy. Excellent for interviews, lectures, and accented speech.

Large

~2.9 GB · Slower

Maximum accuracy. Best for professional work where every word matters.

Turbo

~1.6 GB · Fast

Large-level quality at faster speed. The best all-around choice for most workflows.

Translation

Translate into 20 languages

Powered by a translation model on your own computer — download it once, then translate with no account and no connection.

🇨🇳 Chinese (Simplified) 🇹🇼 Chinese (Traditional) 🇺🇸 English 🇯🇵 Japanese 🇰🇷 Korean 🇫🇷 French 🇩🇪 German 🇪🇸 Spanish 🇮🇹 Italian 🇵🇹 Portuguese 🇷🇺 Russian 🇸🇦 Arabic 🇮🇳 Hindi 🇹🇭 Thai 🇻🇳 Vietnamese 🇮🇩 Indonesian 🇳🇱 Dutch 🇵🇱 Polish 🇸🇪 Swedish

Your audio.
Your machine.
Transcribed.

From file to subtitle in four steps

Everything you need to
work with spoken content.

Your files stay
on your machine.

See the local transcription workflow

Video library

Transcript editor

Offline translation

Export and burn-in

Speech models

Translation engine

Five formats.
One workflow.

Pick your precision

Translate into 20 languages

Learn the workflow

How to Add Subtitles to a Video: A Complete Walkthrough

How to Translate Subtitles Without an API Key

Which Whisper Model Should You Use?

Why Offline Transcription Matters for Confidential Recordings

Start transcribing
today.

Your audio. Your machine. Transcribed.

From file to subtitle in four steps

Everything you need towork with spoken content.

Your files stayon your machine.

See the local transcription workflow

Video library

Transcript editor

Offline translation

Export and burn-in

Speech models

Translation engine

Five formats.One workflow.

Pick your precision

Translate into 20 languages

Learn the workflow

How to Add Subtitles to a Video: A Complete Walkthrough

How to Translate Subtitles Without an API Key

Which Whisper Model Should You Use?

Why Offline Transcription Matters for Confidential Recordings

Start transcribingtoday.

Your audio.
Your machine.
Transcribed.

Everything you need to
work with spoken content.

Your files stay
on your machine.

Five formats.
One workflow.

Start transcribing
today.