Vai al contenuto
AI.info

tools

whisper.cpp

whisper.cpp is a C/C++ implementation for running Whisper speech recognition locally on many platforms.

whisper.cpp

In inglese

whisper.cpp transcribes and translates audio using OpenAI's Whisper model. It provides a command-line tool, C-style API, voice activity detection, real-time microphone examples, model quantization, and CPU or GPU acceleration.

It is used by developers building offline transcription, dictation, subtitle, voice-assistant, and embedded applications. It is an open-source library and local tool, not a hosted transcription service; model files and optional tools such as ffmpeg are obtained separately.

Features

  • C/C++ implementation with a C-style API
  • CPU-only inference and GPU acceleration
  • Apple Silicon support through Metal, Core ML, and Accelerate
  • Integer quantization for smaller model files and lower memory use
  • Voice activity detection through the --vad option
  • Command-line transcription with whisper-cli
  • Real-time microphone transcription example
  • MIT-licensed open-source project

Use cases

  • Transcribe audio files locally without sending recordings to a server
  • Build offline dictation or voice-assistant applications
  • Add speech recognition to iOS, Android, desktop, or embedded software
  • Generate transcripts or subtitles from recorded audio
  • Run speech recognition on CPU, GPU, or Apple Neural Engine hardware

Pros

    Cons

      Latest updates

      • v1.9.4 (v1.9.4)
      • b5127 (b5127)

        Windows arm64 CUDA builds use a preview CUDA Toolkit for Windows on Arm and are experimental.

      • v1.9.2 (v1.9.2)

        Added Parakeet to the XCFramework build, improved Android example inference performance, and updated UTF-8 voice lengths and VAD timestamps.

      • v1.9.0 (v1.9.0)

        Added support for NVIDIA Parakeet and Ruby support for Parakeet.

      • v1.8.7 (v1.8.7)

        The server now merges split UTF-8 token text in verbose JSON; whisper initialization catches C++ exceptions.

      Capabilities

      • Speech to text — “Now build the whisper-cli example and transcribe an audio file like this:” source
      • API — “C-style API” source

      Get it

      Pricing

      Starting price
      $4/user/mo
      Prices checked
      2026-09-25

      Free

      Free

      • Unlimited public/private repositories
      • Dependabot security and version updates
      • 2,000 CI/CD minutes/month
      • 500MB of Packages storage
      • Issues & Projects
      • Community support

      Team

      • $4 USD per user/month
      • Access to GitHub Codespaces
      • Repository rules
      • Multiple reviewers in pull requests
      • Draft pull requests
      • Code owners
      • 3,000 CI/CD minutes/month

      Enterprise

      • Starting at $21 USD per user/month
      • Data residency
      • Enterprise Managed Users
      • User provisioning through SCIM
      • Enterprise Account to centrally manage multiple organizations
      • Environment protection rules
      • Audit Log API
      Official website