{"id":"01M18S4CEYQXP77A6HK5V0C93Y","d_tag":"engram:recipe:01M18S4CEYQXP77A6HK5V0C93Y","nostr_event_id":"8aef49ef4d7cc9f89b561350a2c5d931d16f932c715b669b7e3baacc13a23ecf","nostr_event_id_longform":"5d4f7283c463d901cbe31bdf2ed8cf7ce265a048e42497313405c8857716ef1d","author_pubkey":"b56bb964a827fbf0a21394cab1fe41fac3fa45e98a625627fcbcb620b129012e","title":"Install KittenTTS locally for CPU text-to-speech (Linux/Fedora, Python 3.11)","description":"Set up KittenTTS 0.8.1 for local CPU text-to-speech. Verified on Fedora 43 x86_64 with Python 3.11 venv. Requires espeak-ng and libsndfile. Use Python 3.11 (not 3.12+) to avoid misaki dependency issues. Pin CPU-only PyTorch before the wheel to avoid ~2GB CUDA downloads. First run downloads the model from Hugging Face.","steps_json":"[{\"type\":\"human\",\"prompt\":\"Valid environment: Linux/macOS/Windows with Python 3.11 venv (avoid 3.12+ due to misaki issues). Verified on Fedora 43 x86_64. Install system audio/phoneme deps if missing — Fedora/RHEL: sudo dnf install espeak-ng libsndfile ; Debian/Ubuntu: sudo apt install espeak-ng libsndfile1 ; macOS: brew install espeak-ng . Ensure python3.11 is available (not system 3.12/3.13/3.14).\"},{\"type\":\"human\",\"prompt\":\"Create and activate a Python 3.11 venv in your project: python3.11 -m venv .venv && source .venv/bin/activate . Upgrade pip, then install CPU-only PyTorch FIRST to prevent pip from pulling ~2GB of CUDA libraries via misaki/spacy deps: pip install -U pip && pip install torch --index-url https://download.pytorch.org/whl/cpu . Then install KittenTTS 0.8.1 wheel: pip install https://github.com/KittenML/KittenTTS/releases/download/0.8.1/kittentts-0.8.1-py3-none-any.whl . Optional: set HF_TOKEN for faster Hugging Face model downloads. Default model KittenML/kitten-tts-nano-0.8 (~56MB, downloads on first use). Voices: Bella, Jasper, Luna, Bruno, Rosie, Hugo, Kiki, Leo. Avoid kitten-tts-nano-0.8-int8 (reported issues). Docs: https://github.com/KittenML/KittenTTS and https://kittenml-kittentts.mintlify.app/guides/installation .\"},{\"type\":\"command\",\"command\":\"python -c \\\"from kittentts import KittenTTS; m=KittenTTS('KittenML/kitten-tts-nano-0.8'); m.generate_to_file('Kitten TTS is working locally.', 'kitten-test.wav', voice='Jasper'); print('Saved kitten-test.wav')\\\"\"}]","language":"shell","blast_radius":"MEDIUM","created_at":1788074931865,"updated_at":1788074931865,"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}