LIVE
Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|NVIDIA introduces Nemotron 3.5 Transcribe, a promptable multilingual transcription model28/10/26 · NVIDIA|Failed startups' data finds a second life training AI models28/10/26|Anthropic shares early results on using Claude to accelerate drug development27/10/26 · Anthropic|GEPA-Curator: Filtering Failures to Better Train LLM Agents27/10/26|New Benchmark Tests Whether LLMs Can Uncover Hidden Physical Laws27/10/26|Liquid AI unveils LFM3, a base for specialized multimodal models27/10/26 · Liquid AI|US Police Departments Use AI to Flag 'Suspicious' Driving Behavior27/10/26|Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|NVIDIA introduces Nemotron 3.5 Transcribe, a promptable multilingual transcription model28/10/26 · NVIDIA|Failed startups' data finds a second life training AI models28/10/26|Anthropic shares early results on using Claude to accelerate drug development27/10/26 · Anthropic|GEPA-Curator: Filtering Failures to Better Train LLM Agents27/10/26|New Benchmark Tests Whether LLMs Can Uncover Hidden Physical Laws27/10/26|Liquid AI unveils LFM3, a base for specialized multimodal models27/10/26 · Liquid AI|US Police Departments Use AI to Flag 'Suspicious' Driving Behavior27/10/26|
AppRisingofficial publisher

whisper.cpp

App : A product you use directly

C/C++ implementation of Whisper, very fast on CPU and Apple Silicon.

The engine behind most local transcription apps; real-time use is for technical users.

GitHub · 54,642 ★
Overall score
7.4 /10
Security: nothing to flag
Utility
7.0
Reliability
10.0
Security
No caveat on record
Ease
Technical · Technical

Security reflects where data goes, requested permissions and incidents on record at the verification date. It is not a guarantee of zero risk: check the deployment mode and the connections you enable.

What it does, concretely

Transcribing audio to text usually requires an online service, which means sending potentially sensitive files to third-party servers. whisper.cpp solves this by running OpenAI's Whisper model entirely locally, with no network connection required.

The tool is a C/C++ rewrite of the original Whisper model. It is installed via a few commands, accepts 16-bit WAV audio files, and produces a transcription in the command line. It takes advantage of available hardware acceleration: Apple Neural Engine via Core ML, NVIDIA GPU via CUDA, AMD GPU via ROCm, AMD Ryzen AI NPU via VitisAI, and GPU via Vulkan.

whisper.cpp is aimed at developers and technical users who are comfortable compiling a C++ project and working in a terminal. It also serves as the underlying engine for many local transcription applications that provide a graphical interface on top of it.

It is not suitable for people looking for a ready-to-use clickable interface, or for those who do not want to go through a manual compilation and configuration step.

What it offers

  • ·Fully offline audio transcription, with no data sent to an external server.
  • ·Support for multiple Whisper model sizes: from tiny (75 MB on disk) to large (2.9 GB), depending on the desired speed/accuracy trade-off.
  • ·Acceleration on Apple Silicon via Metal and Core ML, enabling execution on the GPU and Neural Engine.
  • ·NVIDIA GPU, AMD ROCm, Vulkan, and AMD Ryzen AI NPU support to speed up inference on PC.
  • ·Integer quantization of models to reduce memory consumption and disk space.
  • ·Built-in Voice Activity Detection (VAD).
  • ·C-style API available to integrate the engine into other applications or platforms.
  • ·Also runs on iOS, Android, WebAssembly, and Raspberry Pi.

In practice

Price
Free
Commercial use
Commercial use: yes
Available on
cli, macos, windows, linux
Licence
MIT
FR interface
No

Security and data

Rating
Security: nothing to flag
Your data
Data stays local
Maintenance
Active
Last release
6 October 2026

Links and repository checked on 25 October 2026 · added on 1 September 2022

How to install whisper.cpp

1. Clone the repository: git clone https://github.com/ggml-org/whisper.cpp.git 2. Enter the directory: cd whisper.cpp 3. Download a model, for example: sh ./models/download-ggml-model.sh base.en 4. Build the project: cmake -B build && cmake --build build -j --config Release 5. Convert your audio to 16-bit WAV if needed: ffmpeg -i input.mp3 -ar 16000 -ac 1 -c:a pcm_s16le output.wav 6. Run transcription: ./build/bin/whisper-cli -f samples/jfk.wav (make sure to download models only from the official repository)
git clone https://github.com/ggml-org/whisper.cpp.git && cd whisper.cpp && sh ./models/download-ggml-model.sh base.en && cmake -B build && cmake --build build -j --config Release

In the same category

Whole category Audio, voice & music →