Whisper
OpenAI's open-source speech-to-text — transcribe anything locally.
About
Whisper is OpenAI's open-source transcription model — turn any audio or video into accurate text on your own machine, no API needed. The backbone of countless transcription tools; multilingual and robust to accents and noise.
pip install openai-whisper
brew install ffmpeg # macOS — required dependency
sudo apt install ffmpeg # LinuxCopy individual commands
pip install openai-whisperbrew install ffmpeg # macOS — required dependencysudo apt install ffmpeg # Linux
More in Voice & Audio
View allHeartMuLa
Open-source AI song generator — a full song with vocals, or run its 3B model on your own GPU
Voice & AudioFreemium
ElevenLabs
The most realistic AI voices — clone yours in minutes.
Voice & AudioFreemium
Anthropic Agent Skills
Anthropic's official skill collection — documents, spreadsheets, design and more.
Claude SkillsOpen sourceActive189.9K