Frequently Asked Questions (FAQ)
General
What is Subtitle Edit?
Subtitle Edit is a free, open-source subtitle editor. It supports 380+ subtitle formats and provides tools for creating, editing, synchronizing, translating, and converting subtitles.
Is Subtitle Edit free?
Yes. Subtitle Edit is released under the MIT license. It is completely free with no ads or limitations.
Subtitle Edit 5 is built with Avalonia UI and runs on:
- Windows 10 (version 22H2 / build 19045) and later
- Linux (native packages or Flatpak)
- macOS 12 (Monterey) or later, on both Apple Silicon (arm64) and Intel (x64) - macOS 14 (Sonoma) or later recommended, as that is the oldest macOS .NET 10 is supported on
Where is my data stored?
Subtitle Edit stores settings and data in a platform-specific data folder. You can open it with the keyboard shortcut Ctrl+Alt+Shift+D (Windows/Linux) or Cmd+Alt+Shift+D (macOS).
How do I report a bug or request a feature?
Please open an issue on the GitHub repository.
File Operations
Subtitle Edit supports over 380 subtitle formats including SubRip (.srt), WebVTT (.vtt), Advanced SubStation Alpha (.ass), SubStation Alpha (.ssa), MicroDVD, SAMI, and many more. See Supported Formats for the full list.
- Open the subtitle file via File → Open
- Choose a new format from the format dropdown
- Save via File → Save as…
For batch conversion, use Tools → Batch convert….
How do I import image-based subtitles?
Use File → Import → Image-based subtitle for OCR… to import Blu-ray SUP, VobSub, or other image-based subtitles and convert them to text using OCR.
You can use File → Import → Image-based subtitle for edit… to work/export the images.
Video
What video players are supported?
Subtitle Edit supports:
- libmpv — High-quality, wide format support (default). Several rendering modes: OpenGL, Native Window ID (not macOS), Software rendering, and on macOS libmpv - Metal (experimental, needs restart)
- libVLC — Popular player, limited support in SE (Windows and Linux only)
- FFmpeg — Plays video through FFmpeg (all platforms)
Configure the video player in Options → Settings → Video player.
If you have problems with a video, try Video → More → Re-encode video for better subtitling.
Open a video file, and Subtitle Edit will generate a waveform automatically if FFmpeg is installed.
To enable the spectrogram, go to Options → Settings.
Where do I get FFmpeg?
Some functions in Subtitle Edit will prompt for downloading FFmpeg.
You can also download FFmpeg in Options → Settings.
Speech to Text
What speech-to-text engines are supported?
Subtitle Edit supports several local and downloadable speech-to-text engines:
- Whisper CPP — Cross-platform, CPU-based. On Windows the cuBLAS (NVIDIA) and Vulkan GPU backends are picked from its Backend drop-down
- Purfview Faster Whisper XXL — Fast GPU/CPU-based (Windows, Linux x64)
- Whisper CTranslate2 — Fast CPU/GPU-based
- Whisper Const-me — DirectX-based (Windows)
- WhisperX — Faster-Whisper with wav2vec2 word-level alignment, standalone build (Windows x64, Linux x64, macOS Apple Silicon)
- Whisper OpenAI — Original Python implementation
- Qwen3 ASR CPP and Crisp ASR (many backends: Parakeet, Canary, Cohere, Qwen3 and more) — local engines
- OpenAI Compatible Server, OpenRouter, Alibaba Qwen3-ASR and Google Cloud Speech-to-Text — online services
See Speech to Text for details.
- OpenAI-compatible STT — Send audio to any OpenAI-compatible speech endpoint
- Qwen3 ASR CPP — Local GGUF-based engine (Windows, Linux, macOS)
- Crisp ASR — Local engine with multiple backends (Parakeet, Canary, Cohere, Fire Red, Fun-ASR Nano, GigaAM, GLM, Granite, Qwen3, Mega, MOSS Diarize, Omni, Kyutai, SenseVoice, ARK, Voxtral) and a Forced-aligner picker (built-in / Canary CTC / Qwen3 / 12 wav2vec2 language aligners) for word-level timestamps
How do I use Speech to Text?
- Open a video file
- Go to Video → Speech to text…
- Choose an engine and model (download if needed)
- Select the language
- Click Transcribe
See Speech to Text for details.
Synchronization
How do I fix subtitle timing that is off by a constant amount?
Use Synchronization → Adjust all times… and enter the offset in milliseconds.
How do I sync subtitles that drift over time?
Use Synchronization → Visual sync… or Synchronization → Point sync… to synchronize at multiple points.
How do I convert subtitles between different frame rates?
Use Synchronization → Change frame rate… and select the source and target frame rates.
OCR
What OCR engines are available?
- Tesseract — Open-source OCR engine with language packs
- nOCR — Nikse’s built-in OCR with trainable databases
- Binary OCR — For binary image comparison
- PaddleOCR — Local OCR (standalone build for Windows/Linux, or the pip-installed Python engine)
- CrispEmbed — Local OCR with downloadable model backends (PP-OCRv6, GLM-OCR, GOT-OCR2, Qwen3-VL, DeepSeek-OCR-2)
- llama.cpp — Local vision-LLM OCR with a managed server and curated models
- Ollama OCR — Local LLM-based
- Apple Vision — Built-in OCR on macOS
- Google Lens OCR — Cloud-based
- Google Vision OCR — Cloud-based
- Mistral OCR — Cloud-based
How do I OCR Blu-ray subtitles?
- Open the
.sup file via File → Open
- The OCR window will open automatically
- Choose an OCR engine (Tesseract or nOCR recommended)
- Start the OCR process
Spell Check
How do I install spell check dictionaries?
Go to Spell check → Get dictionaries… and download the dictionary for your language. Dictionaries are based on Hunspell/OpenOffice format.
How do I add words to the dictionary?
During spell check, click Add to user dictionary to add a word. You can also manage custom words in Options → Word lists.
ASSA / Styling
What is ASSA?
ASSA (Advanced SubStation Alpha) is a subtitle format that supports rich styling including fonts, colors, positioning, animations, and more. It is widely used in anime fansubbing and professional workflows.
How do I edit ASSA styles?
Go to ASSA tools → Styles… to create and edit subtitle styles with fonts, colors, borders, shadows, and alignment.
Keyboard Shortcuts
How do I customize keyboard shortcuts?
Go to Options → Shortcuts… to view and customize all keyboard shortcuts.
What are the default shortcuts?
See the Keyboard Shortcuts Reference for the complete default shortcut list.
Troubleshooting
Subtitle Edit won’t play video
- Ensure you have a video player installed (libmpv or VLC)
- Check that the video player path is correctly set in Options → Settings → Video player
- Try to re-encode the video for better compatibility via Video → More → Re-encode video for better subtitling
The waveform is made with FFmpeg while the picture comes from libmpv, so this points at libmpv failing to load rather than at the file.
- Update your graphics driver from Intel/NVIDIA/AMD directly — the generic Microsoft Basic Display driver has no usable OpenGL
- Re-download libmpv via Options → Settings → Video player → Download libmpv
- If you installed libmpv by hand, make sure
vulkan-1.dll sits next to libmpv-2.dll — recent libmpv builds do not load without it (see Third-Party Components)
- Check that your antivirus is not blocking
libmpv-2.dll
- As a cross-check, switch to VLC under Options → Settings → Video player; if VLC plays the file, the problem is libmpv/OpenGL
Video does not play on Linux
With the native .tar.gz builds, Subtitle Edit uses the system libmpv. If it is not installed (common on Fedora), the video area stays empty, with no picture or sound.
- Install libmpv and FFmpeg with your package manager, e.g.
sudo apt install libmpv-dev ffmpeg (Debian/Ubuntu) or sudo dnf install mpv-libs ffmpeg-free (Fedora) - see the README for other distributions
- Or use the Flatpak, which bundles mpv and FFmpeg
- As a fallback, switch to FFmpeg under Options → Settings → Video player
Video does not play on macOS
The .dmg bundles libmpv and FFmpeg, so nothing needs to be installed. If video still does not show, try another player under Options → Settings → Video player (e.g. FFmpeg, or libmpv - Metal (experimental, needs restart) - restart Subtitle Edit after switching to it).
- Ensure FFmpeg is installed and the path is set in settings
Speech to Text fails
- Ensure the Whisper engine is downloaded (it will prompt you)
- Ensure the model is downloaded
- Check the console log in the Speech to Text window for error details
- Try a smaller model if you run out of GPU memory