Embeddable Speech/Audio stable

Whisper.cpp

Port of OpenAI Whisper speech recognition to C/C++

37.0K stars 100 contributors Since 2022
Website → GitHub

Port of OpenAI Whisper speech recognition to C/C++

License
MIT
Min RAM
512 MB
Min CPUs
1 core
Scaling
single_node
Complexity
intermediate
Performance
medium
Self-hostable
✓
K8s native
✕
Offline
✓
Pricing
fully free
Docs quality
good
Vendor lock-in
none

Use cases

  • ✓ Local speech-to-text without API calls
  • ✓ Generate subtitles from video/audio
  • ✓ Voice input for applications
  • ✓ Meeting transcription

Anti-patterns / when NOT to use

  • ✕ Accuracy depends on model size
  • ✕ Real-time requires GPU for large models
  • ✕ English-centric accuracy for smaller models

Replaces / alternatives to

  • OpenAI Whisper API
  • Google Speech-to-Text
  • AWS Transcribe

Technical specs

Language
CC++
API type
SDK
Protocols
HTTP
Deployment
sourcebinary
SDKs
cc++pythonjavascriptrustgo

Community

GitHub stars 37.0K
Contributors 100
Commit frequency weekly
Plugin ecosystem none
Backing Georgi Gerganov
Funding community

Release

Latest version —
Last release —
Since 2022

Best fit

Team size
solosmallmedium
Industries
general

Tags

  • speech-to-text
  • whisper
  • transcription
  • local
  • cpu
  • gpu
  • wasm
  • multilingual
  • real-time