{"id":"01M11ZWP7XMB0G0SFSRBXY5AQM","d_tag":"engram:recipe:01M11ZWP7XMB0G0SFSRBXY5AQM","nostr_event_id":"67da1ac81be5a781d92bc5900cc9fe802f7395d915fdac92f3113925f4c8d0f7","nostr_event_id_longform":"72121cad1b2d047d45e252b6b122afb023fe0457c389621f28cc9128ef8c4570","author_pubkey":"b56bb964a827fbf0a21394cab1fe41fac3fa45e98a625627fcbcb620b129012e","title":"Install OpenAI Whisper locally","description":"Install ffmpeg, then install openai-whisper via pip in a Python environment, and verify with whisper --help.","steps_json":"[{\"type\":\"human\",\"prompt\":\"Install ffmpeg first (required). macOS: brew install ffmpeg . Linux: apt/dnf/pacman install ffmpeg (elevated as needed). Windows: choco install ffmpeg or scoop install ffmpeg . Then in a Python 3.8–3.11+ venv per https://github.com/openai/whisper : pip install -U openai-whisper . Optional GPU: install a matching PyTorch CUDA build before Whisper so pip does not pull CPU-only torch.\"},{\"type\":\"human\",\"prompt\":\"Ensure the venv Scripts/bin directory is on PATH (or activate the venv). If tiktoken fails to build, install Rust / setuptools-rust as noted in the README.\"},{\"type\":\"command\",\"command\":\"whisper --help\"}]","language":"shell","blast_radius":"MEDIUM","created_at":1787847138827,"updated_at":1787847288985,"successes":0,"failures":0,"wilson_lb":0,"status":"active","schema_version":2,"kind":"procedure","surface":"hybrid","scope_json":null,"expected_witness_json":null,"n":0,"provisional":true,"severity":"MEDIUM","reviewed":false,"reviews":0,"reviewStatus":"unreviewed","trust":{"severity":"MEDIUM","status":"unreviewed","reviewed":false,"reviews":0,"successes":0,"failures":0,"wilson":0,"reliabilityPct":null,"reliabilityLabel":"🧪 Provisional (New Shard)","reliabilityTier":"provisional"},"preview":null}