NeMo Parakeet ASR Models attain strong speech recognition accuracy while being efficient for inference. Available in CTC and RNN-Transducer variants.
whisper.cpp
The whisper.cpp project ports OpenAI’s Whisper speech recognition model to C and C++.
About whisper.cpp
whisper.cpp is a C/C++ port of OpenAI's Whisper automatic speech recognition model, designed for high-performance inference without external dependencies. It can run on CPU-only systems or use GPU and NPU backends, and it supports a range of platforms including macOS, iOS, Android, Linux, Windows, WebAssembly, Raspberry Pi and Docker. The repository includes command-line examples such as whisper-cli, which currently accepts only 16-bit WAV files.
Key features
- Plain C/C++ implementation without external dependencies
- Supports CPU-only inference and GPU backends including NVIDIA, AMD ROCm and Vulkan
- Apple Silicon optimized via ARM NEON, Accelerate framework, Metal and Core ML
- Integer quantization and mixed F16/F32 precision for lower memory and disk use
- C-style API and Voice Activity Detection
- Runs on macOS, iOS, Android, Linux, Windows, WebAssembly, Raspberry Pi and Docker
Best for: Developers who want to integrate offline Whisper speech recognition into C/C++ applications across desktop, mobile or embedded platforms.
Good for
- Open source and lightweight, making it easy to integrate into different platforms and applications.
- Supports a wide range of hardware backends, including CPU-only, Apple Silicon, NVIDIA, AMD ROCm, Vulkan and several NPUs.
- Includes example command-line tools for quick transcription and model downloads.
Watch out for
- The whisper-cli example currently runs only with 16-bit WAV files, so other formats must be converted first.
- Memory and disk requirements grow with model size, from about 273 MB to 3.9 GB of memory for tiny to large models.
whisper.cpp at a glance
- Category
- AI & Machine Learning › AI Voice & Audio
- Type
- Open-source project
- Pricing
- Not confirmed. Check the site
- Open source
- Yes · Source code
- Language
- English
- Last checked
- Oct 6, 2026 · Online when checked
- Listed
- Oct 5, 2026
whisper.cpp alternatives
Tools and sites you could use instead, ranked by how closely they match.
The wav2letter project provides an automatic speech recognition toolkit from Facebook AI Research.
Superwhisper turns speech into polished text for use in apps such as Slack and Gmail, with recognition options listed for four operating systems.
The whisper-ctranslate2 command-line client works with the OpenAI Whisper client and uses CTranslate2.
Open source
24 alternatives
Vosk provides an offline speech recognition API for mobile devices and servers, with several programming language interfaces.
Open source
24 alternatives
ESPnet includes tools for end to end speech processing and related speech tasks.
Open source
24 alternativesPopular in AI Voice & Audio
Well-known listings from the same topic.
Google Illuminateilluminate.google.comAlteredaltered.aiELSA Speakelsaspeak.comKreadoAIkreadoai.comPodwisepodwise.aiAddsubtitleaddsubtitle.ai
whisper.cpp FAQ
What audio format does the whisper-cli example accept?
It currently runs only with 16-bit WAV files; other formats must be converted, for example with ffmpeg.
Can whisper.cpp run without a GPU?
Yes, it supports CPU-only inference.
How much memory do the models require?
Approximate memory usage ranges from about 273 MB for tiny to about 3.9 GB for large, with disk usage from 75 MiB to 2.9 GiB.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.