Subtitle Edit

the subtitle editor :)

Speech to Text

Subtitle Edit can automatically transcribe audio to text using Whisper-based and other modern speech recognition engines.

Menu: Video → Speech to text…

Speech to Text

Supported Engines

Engine	Platform	Notes
Whisper CPP	Windows, Linux, macOS	Local CPU engine. On Windows the cuBLAS (NVIDIA CUDA) and Vulkan GPU backends can also be selected from the Whisper CPP backend dropdown.
Purfview Faster Whisper XXL	Windows, Linux	Fast local engine, often used with NVIDIA CUDA
Whisper CTranslate2	Windows, Linux (x64), macOS (Apple Silicon)	CPU / NVIDIA CUDA depending on installation; CUDA requires CUDA 12.x
MLX Whisper	macOS (Apple Silicon)	Runs Whisper on the Apple GPU / Neural Engine via Apple’s MLX. Not downloaded by Subtitle Edit — install the `mlx-whisper` Python package yourself (see notes below)
Whisper Const-me	Windows	DirectX-based engine
Whisper OpenAI	All	Python-based OpenAI Whisper workflow
OpenAI Compatible Server	All	Connect to any OpenAI-compatible speech-to-text endpoint
OpenRouter	All	Online. One API key routes to Whisper, gpt-4o-transcribe, Groq and Google Chirp
Alibaba Qwen3-ASR	All	Online Qwen3-ASR via Alibaba Model Studio (DashScope)
Qwen3 ASR CPP	Windows, Linux	Local Qwen3 ASR engine with downloadable GGUF models
Crisp ASR	Windows, Linux, macOS	Single engine with selectable backends: Parakeet, Canary, Cohere, Fire Red, Fun-ASR Nano, Fun-ASR MLT Nano, GigaAM, GLM, Granite, Qwen3, Mega, MOSS Diarize, Omni, Kyutai, SenseVoice, ARK, Voxtral

Engines and models are downloaded automatically on first use.

SE5 Engine Notes

Whisper CPP is shown as a single entry; the CPU / cuBLAS / Vulkan backends are selected from a secondary dropdown when Whisper CPP is selected.
Qwen3 ASR CPP includes 0.6B and 1.7B model options, plus a forced-aligner model used for timing workflows.
Crisp ASR is exposed as one engine that wraps multiple backends (Parakeet, Canary, Cohere, Fire Red, Fun-ASR Nano, Fun-ASR MLT Nano, GigaAM, GLM, Granite, Qwen3, Mega, MOSS Diarize, Omni, Kyutai, SenseVoice, ARK, Voxtral). Pick the backend from the Crisp ASR backend dropdown - see Crisp ASR backends for what each one is good at.
MLX Whisper (Apple Silicon Macs) is not bundled or auto-downloaded — it drives Apple’s mlx-whisper Python package. Install it once with pip3 install mlx-whisper (or pipx install mlx-whisper); models download from Hugging Face on first use. Subtitle Edit detects the install by finding a Python that can import mlx_whisper — it probes Homebrew, python.org, pyenv and system interpreters, and (for pipx / virtual-env / conda installs, which isolate the package) reads the mlx_whisper command found on your PATH or at ~/.local/bin/mlx_whisper to locate the matching interpreter. If it reports “not found” after a pipx/venv install, make sure which mlx_whisper resolves.
A Forced aligner option is shown for Crisp ASR backends and exposes the built-in aligner, Canary CTC, Qwen3, and the wav2vec2 zoo (12 language-specific CTC aligners that run on top of any Crisp ASR backend).
Several newer engines support automatic language selection.
Each engine can have separate advanced command-line parameters.

Crisp ASR backends

Crisp ASR is one engine wrapping many backends. Set Engine to Crisp ASR, then pick the backend from the Crisp ASR backend dropdown - the model list updates to that backend’s models.

The languages column counts what the backend dropdown offers (an auto entry is not counted as a language).

Output is what you actually get back. Native timings means the backend times its own lines and can use the built-in aligner; needs aligner means the text arrives untimed and a forced aligner (Canary CTC, Qwen3, or a wav2vec2 aligner) places it in time. Anything else a backend adds - speaker labels, punctuation and casing - is called out there too.

Backend	Languages	Output	Model size range	Good for
Parakeet	13 (European + zh, ja, ko)	Native timings	75 MB - 2.14 GB	The fast default. NVIDIA Parakeet TDT/RNN-T in 0.6B, 1.1B and a 110M tdt_ctc model - the smallest Crisp ASR model of all. Has a Japanese fine-tune
Canary	25 (European)	Native timings	705 MB - 1.97 GB	NVIDIA Canary 1B v2. Broad European coverage with its own timings; also usable as a CTC forced aligner for other backends
Cohere	14	Native timings	1.51 - 4.14 GB	Cohere Transcribe. Separate Arabic and Japanese fine-tunes. VAD is on by default (see the VAD section below)
GigaAM	1 (Russian)	Native timings; punctuation + casing on `e2e` only	151 - 452 MB	Russian only, and very small. Use an `e2e` revision - those emit punctuation and capitalisation, the plain `ctc` / `rnnt` heads return bare lowercase text
MOSS Diarize	2 (en, zh)	Native timings + speaker labels	1.41 - 1.82 GB	Labels who is speaking. This is the backend preselected by the text-to-speech speaker-separation workflow
Qwen3	30	Needs aligner	631 MB - 4.7 GB	Strong all-rounder in 0.6B and 1.7B, with an anime / visual-novel Japanese fine-tune. Also supplies the Qwen3 forced-aligner model
Omni	149	Needs aligner	1.08 - 3.26 GB	By far the widest language coverage - the place to start for a language no other backend lists. Languages use NLLB-style codes (`eng_Latn`, `cmn_Hans`)
Fire Red	49	Needs aligner	1 - 2.4 GB	FireRedASR2. Chinese-first, and the only backend covering Chinese regional languages: Cantonese, Shanghainese, Minnan, Gan, Hakka and Xiang
GLM	17	Needs aligner	1.3 - 4.5 GB	GLM-ASR Nano, Chinese-first with a wide second tier of languages
SenseVoice	5 (zh, yue, en, ja, ko)	Needs aligner	136 - 469 MB	Tiny and quick for CJK audio - useful on machines where the larger backends are too slow
Fun-ASR Nano	5 (en, zh, yue, ja, ko)	Needs aligner	0.90 - 1.98 GB	CJK-focused Fun-ASR
Fun-ASR MLT Nano	31	Needs aligner	0.90 - 1.98 GB	The multilingual sibling of Fun-ASR Nano at the same size
Mega	2 (en, zh)	Needs aligner	1.3 - 4.4 GB	Mega-ASR 1.7B. VAD is on by default (see the VAD section below)
Granite	6 (en, fr, de, es, pt, ja)	Needs aligner	1.54 - 5.58 GB	IBM Granite Speech 4.1 2B. The `plus` models are the newer revision; `mini` and `f16enc` trade encoder precision for size
ARK	19 (European + zh, ja, ko)	Needs aligner	3.52 - 7.51 GB	A 3B model - the heaviest backend here, so only worth it when the smaller ones fall short
Kyutai	2 (en, fr)	Needs aligner	0.67 - 5.01 GB	Kyutai STT in 1B and 2.6B
Voxtral	8	Needs aligner	2.65 - 4.99 GB	Mistral Voxtral Mini 3B. The backend has no built-in aligner entry at all, so a CTC aligner is always used

Picking a quantization

Most backends list the same model several times with a quantization suffix. The suffix only changes file size, memory use and speed - the model is the same:

q4_k - smallest and fastest, with the most accuracy lost. A good first download
q5_0 / q5_1 / q6_k - middle ground where offered
q8_0 - close to full precision at roughly half the size. The best default when disk space allows
no suffix / f16 / unquantized - full precision, largest and slowest, and rarely worth it over q8_0

A model name with a language in it (-ja, -arabic) is a fine-tune for that language and usually beats the general model on it - see below.

Language-specific models

Some backends offer models fine-tuned for one language, which usually beat the general model on that language.

These models live under the Crisp ASR engine, not under the standalone Qwen3 ASR CPP engine. To reach them, set Engine to Crisp ASR, then pick the backend from the Crisp ASR backend dropdown - the model list updates to that backend’s models. For Japanese:

Backend	Model	Notes
Crisp ASR Qwen3	`qwen3-asr-1.7b-ja-anime-q8_0.gguf` (or `-q4_k.gguf`)	Fine-tuned on anime / visual novel speech - the best starting point for anime audio
Crisp ASR Cohere	`cohere-asr-ja-q8_0.gguf` (also q4_k / q6_k / f16)	Japanese fine-tune covering general and anime domains
Crisp ASR Parakeet	`parakeet-tdt-0.6b-ja-q8_0.gguf` (also q4_k / unquantized)	Fast Japanese model

The general qwen3-asr-1.7b and Whisper large-v3-turbo models are also strong on Japanese if you prefer a single model for mixed content.

How to Use

Open a video file in Subtitle Edit
Go to Video → Speech to text…
Select an Engine from the dropdown
Select a Model (larger models usually improve accuracy but take more time and disk space)
Select the Language of the audio, or use auto-language when the selected engine supports it
Optionally enable:
- Translate to English — Translate non-English audio to English
- Adjust timings — Post-process timing using waveform data
- Post-processing — Fix casing, merge lines, add periods, etc.
Click Transcribe

Models

Each engine has its own set of models. Common model sizes:

tiny — Fastest, least accurate
base — Good balance for quick work
small — Better accuracy
medium — High accuracy
large / large-v2 / large-v3 — Best accuracy, slowest

Models ending in .en are English-only and perform better for English audio.

Batch Mode

Transcribe multiple video files at once:

Click Batch mode
Add video files
Click Transcribe
Results are saved as .srt files next to the video files

Advanced Settings

Click the Advanced button to configure custom command-line arguments for the Whisper engine:

Use VAD (voice activity detection) for better timing
Highlight spoken words in the transcript
Adjust temperature or other model parameters

Advanced settings are stored per engine, so you can keep separate parameters for Whisper CPP, Qwen3 ASR, Crisp ASR, and other engines.

Voice activity detection with Crisp ASR Cohere and Mega

For the Cohere and Mega backends, Subtitle Edit adds --vad and the bundled Silero VAD model to the command line by default. Crisp ASR turns VAD on for these models by itself on longer audio, so passing the bundled model mainly keeps Crisp ASR from downloading its own copy in the middle of a transcription.

VAD usually gives tighter timings, but on some material it drops quiet speech and clips the first word of a line. Crisp ASR has no --no-vad switch, so VAD is turned off by asking for fixed chunking instead: add --chunk-seconds 30 (or -ck 30) to the advanced parameters, and Subtitle Edit leaves --vad out of the command line entirely. Expect the trade-off that comes with it - with VAD off, long stretches of silence or music can produce invented lines.

To keep VAD and only change how it behaves, put --vad in the advanced parameters yourself (the Enable VAD button fills in the flag and the model path). An explicit --vad wins over --chunk-seconds, so the two can be combined.

Post-Processing Settings

Click the Post-processing button to configure:

Adjust timings (using waveform peak data)
Fix short durations
Fix casing
Add periods
Merge short lines
Split long lines
Change underline to color (useful for highlight spoken words)

Console Log

The console log at the bottom shows real-time output from the Whisper process, useful for debugging issues.

Tips

For NVIDIA GPU users, use the Whisper CPP cuBLAS backend or Purfview Faster Whisper XXL for fastest transcription
If you get “CUDA out of memory” errors, try a smaller model
The --standard parameter is automatically added for Purfview Faster Whisper XXL
You can re-download an engine by right-clicking the engine area
If an engine executable has gone missing (typically quarantined by antivirus software), Subtitle Edit detects it when transcription starts, names the missing file, and offers to re-download the engine
If a new engine has no model installed yet, let Subtitle Edit download both the engine and the selected model before starting transcription