whisper-ctranslate2

What is whisper-ctranslate2?

whisper-ctranslate2 is an open-source command-line program for transcribing and translating speech with Whisper models. It follows the same command-line options as OpenAI's own Whisper client, so existing scripts can switch over with little change, but it runs on the CTranslate2 engine and the faster-whisper implementation. The project reports about four times the speed of the original at the same accuracy with lower memory use, and up to sixteen times faster with batched inference. Extra options add speaker diarization through pyannote.audio, a Silero-based voice activity detection filter, int8 and other quantization types, your own fine-tuned or converted models, colored confidence output and live transcription from a microphone. You install it with pip or pull a Docker image that includes the small, medium and large-v2 models. It runs on x86-64 and ARM64 CPUs and on NVIDIA GPUs, and there is no price to use it.

Use cases

• Transcribe audio files from the command line
• Translate speech to English text
• Identify speakers in a recording
• Transcribe live from a microphone
• Migrate scripts from OpenAI's Whisper CLI
• Filter silence with voice activity detection
• Run fine-tuned Whisper models locally

Who it's for

• Developers
• Researchers
• Data engineers
• Podcast and media teams

Why use whisper-ctranslate2

• Transcription is several times faster than the original Whisper at the same accuracy, according to the project.
• Because the options match OpenAI's CLI, moving existing workflows over takes little effort.
• Runs on your own CPU or GPU, so audio stays on your machine and there are no usage fees.
• Built-in diarization, VAD and live mode cover common needs without extra tooling.
• Bring your own quantized or fine-tuned model through the model directory option.

whisper-ctranslate2 features

• Command line compatible with the OpenAI Whisper client
• CTranslate2 and faster-whisper engine
• Batched inference for extra speed
• Speaker diarization using pyannote.audio
• Voice activity detection filter based on Silero VAD
• Quantization options such as int8 and float16
• Load your own CTranslate2 or fine-tuned Whisper models
• Live transcription from the microphone
• Color-coded word confidence output
• Docker image with preloaded models

whisper-ctranslate2 integrations

• CTranslate2
• faster-whisper
• pyannote.audio
• Silero VAD
• Hugging Face

How to get started with whisper-ctranslate2

1. Install with pip install whisper-ctranslate2, or pull the Docker image.
2. Run whisper-ctranslate2 followed by your audio file and a model name such as medium.
3. Add options like --task translate, --vad_filter True or --batched True for translation, silence filtering or extra speed.
4. For speaker labels, install pyannote.audio, accept the model conditions and pass a Hugging Face token with --hf_token.
5. Collect output files such as JSON, VTT and SRT, where speaker names appear when diarization is on.

whisper-ctranslate2 limitations

• Translation targets English only
• Diarization works at segment level and needs a Hugging Face access token plus accepted model conditions
• GPU execution needs NVIDIA cuBLAS and cuDNN for CUDA 12
• Batched inference ignores several decoding parameters such as no_speech_threshold and condition_on_previous_text
• The Docker image bundles only the small, medium and large-v2 models

whisper-ctranslate2 pricing

whisper-ctranslate2 is open source, so you can use and self-host it for free.

Prices can change and may vary by country or exclude tax. Check the official pricing page.

whisper-ctranslate2 alternatives

The main whisper-ctranslate2 alternatives listed on Debuly are whisper.cpp, WhisperX, Deepgram, CAMB.AI, Qwen3-TTS and TTS WebUI. The table compares pricing and platforms; order is A–Z.

1. 

   CAMB.AI

   AI dubbing, text-to-speech, translation and voice cloning across 140+ languages, with a studio app and developer API.

   Audio and voiceVideoDeveloper platforms
   FreemiumVisit CAMB.AI (official site)
2. 

   Deepgram

   Voice AI platform with speech-to-text, text-to-speech and Voice Agent APIs, available as a managed API, in-VPC or self-hosted.

   Audio and voiceDeveloper platformsAutomation and agents
   FreemiumVisit Deepgram (official site)
3. 

   Qwen3-TTS

   Open-source text-to-speech model family from Qwen with voice cloning, voice design from a text description and low-latency streaming.

   Audio and voiceDeveloper platforms
   Open sourceVisit Qwen3-TTS (official site)
4. TW

   TTS WebUI

   Open-source Gradio and React interface that runs many text-to-speech, music and audio models locally through extensions.

   Audio and voiceDeveloper platforms
   Open sourceVisit TTS WebUI (official site)
5. 

   whisper.cpp

   Dependency-free C/C++ port of OpenAI's Whisper speech recognition that runs offline on CPUs, Apple Silicon and many GPUs.

   Audio and voiceDeveloper platforms
   Open sourceVisit whisper.cpp (official site)
6. WX

   WhisperX

   Open-source speech recognition built on Whisper, adding word-level timestamps, speaker diarization and fast batched transcription.

   Audio and voiceDeveloper platforms
   Open sourceVisit WhisperX (official site)

whisper-ctranslate2 FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: from avatars.githubusercontent.com · fetched 24 Oct 2026 · Vendor trademark, used only to identify the product; removed on request
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.