Show HN: Stable Audio 3 – one-shot sample generator (110gb download)

Stability AI released Stable Audio 3, a one-shot sample generator that requires an 110GB download, with the v3 libraries hosted publicly on Google Cloud Platform without authentication. The tool, part of the Signals & Sorcery app, allows users to generate large batches of drum and pitched instrument audio samples using rented GPU power from RunPod, with a full 50-category campaign costing under $50 and taking up to a day to complete.

The latest v3 libraries are hosted publicly on GCP no auth required . The Signals & Sorcery https://signalsandsorcery.com app installs these automatically, but you can grab them directly here: | Pack | Contents | Download | |---|---|---| Drums v3 large | 24 roles · 10,359 one-shots + prompt sidecars · ~1.4 GB | | Instruments v3 large sas-instrument-pack-v3-large.zip https://storage.googleapis.com/docs-assets/sas-instrument-pack-v3-large.zip Each zip contains a pack-version.json marker plus the payload tree drums: <role / .wav ; instruments: <category /<id /manifest.json + zones/ . The instrument pack is zones-only the 24-bit generation sources/ are omitted — they aren't used at playback . Generate large batches of audio samples with Stable Audio 3 https://huggingface.co/stabilityai/stable-audio-3-medium on a rented RunPod https://www.runpod.io GPU. Two pipelines ship side by side: Drums / one-shots run all.sh — 24 unpitched categories kicks, snares, hats, claps, 808s, risers, impacts, textures… . Generate → quality-gate → trim/normalize → flat processed/<role / folders. Pitched instruments run pitched.sh — 28 tuned categories pianos, basses, pads, leads, strings… . Generate → pitch/quality-gate → multi-source pitch-correct + pre-render playable zones → instruments/<cat /<id /manifest.json . Both run the same retry-to-target loop re-roll failed prompts until ~150 samples per category survive the gate and batched generation on one model load. Read top to bottom, copy-paste each command block. A single-category test slice is ~5 minutes and a few cents; the full ~50-category v3 campaign is up to ~a day of big-GPU pod time still well under $50 — see Cost cost-recap . Assumes an Apple Silicon Mac as the control machine. For the rationale why these settings, prompt-design tips, deep cost math , see stable audio open batch oneshot guide.md /shiehn/sas-sample-generator/blob/main/stable audio open batch oneshot guide.md . Part of the Signals & Sorcery family. See Related repos at the bottom of this README. Plain-text prompt files — one description per line — under prompts/ drums and prompts/pitched/ instruments . The repo ships 52 categories, 200 prompts each ~10,400 prompts , pre-generated by scripts/gen prompts.py /shiehn/sas-sample-generator/blob/main/scripts/gen prompts.py in a combinatorial house style ~58% EDM / ~25% hip-hop & urban / ~17% acoustic-orchestral-world . Run them as-is, edit them, or subset which categories generate. Each non-comment line becomes one generation job. Example from prompts/kick.txt /shiehn/sas-sample-generator/blob/main/prompts/kick.txt : 909-style tight 909-style kick drum one shot, hard click transient, short punchy body, dry punchy 909 kick drum one shot, sharp transient, controlled low end, clean studio sample 808-style deep 808 kick one shot, long sub bass decay, smooth sine low end, dry warm 808 kick one shot, saturated low end, medium decay, dry, no melody, no loop Blank lines and lines starting with are ignored handy for grouping . Aim for ~10 words per line. For drums, always include phrases like one shot, no loop so the model doesn't render a rhythmic loop. Editing prompts changes content-hashes.Output filenames are content-addressed {category}-{hash}.wav , so re-wording a line orphans the WAV it used to produce. Finalize wordingbeforea GPU run. Re-running with --skip-existing then only generates the new/changed lines. Flat processed/<role / output; the folder name is the drum role. | Core kit | Dur | EDM / electronic one-shots | Dur | | |---|---|---|---|---| kick | 1.5s | clap | 0.75s | | snare-standard | 1.0s | 808 tuned sub one-shot | 2.0s | | snare-rim | 0.75s | riser | 4.0s | | hat-closed | 0.5s | downlifter | 3.0s | | hat-open | 1.5s | impact | 2.0s | | cymbal-ride | 2.5s | sub-drop | 2.0s | | cymbal-crash | 3.0s | sweep | 2.5s | | cymbal-splash | 1.5s | texture vinyl/foley/glitch | 3.0s | | tamborine | 1.0s | zap | 0.75s | | shaker | 0.75s | foley-perc | 0.75s | | tom-hi / tom-mid / tom-low | 1.0–1.5s | ||| hit generic stab | 1.5s | instruments/<cat /<id / with a manifest.json . Source pitches, durations, and variant counts are in the Pitched-instrument pipeline pitched-instrument-pipeline table below. synths lead-supersaw lead-fm lead-acid pluck-synth plucks keys pianos organs basses 808-bass reese-bass pads strings brass winds accordion bells mallets percussion timpani guitars banjos mandolin harp sitar vocals choir Output filenames are content-addressed: {category}-{hash}.wav . Same prompt + seed → same filename → safely re-runnable with --skip-existing . To subset what generates, edit scripts/categories.txt /shiehn/sas-sample-generator/blob/main/scripts/categories.txt drums or pitched — comment out any line to skip that category. /shiehn/sas-sample-generator/blob/main/scripts/pitched categories.txt scripts/pitched categories.txt - Create / sign in at huggingface.co https://huggingface.co . - Visit stabilityai/stable-audio-3-medium https://huggingface.co/stabilityai/stable-audio-3-medium and click Agree and access repository . SA3 also requires accepting the Gemma Terms of Use linked from the same page . If you switch models via --model , accept the license on that model's page too — stable-audio-3-small-sfx https://huggingface.co/stabilityai/stable-audio-3-small-sfx is the lighter 0.6B SFX-tuned alternative. - Create a read-only access token at huggingface.co/settings/tokens https://huggingface.co/settings/tokens . Save it in your password manager — you'll paste it once per pod. - Create / sign in at runpod.io https://runpod.io . Add a payment method. - Add your Mac SSH public key under Settings → SSH Keys https://www.runpod.io/console/user/settings :Paste it into the form. If pbcopy < ~/.ssh/id ed25519.pub copies key to clipboard ~/.ssh/id ed25519.pub doesn't exist: ssh-keygen -t ed25519 first, accept defaults. runpod.io/console/pods https://www.runpod.io/console/pods → Deploy → GPU Pod : | Setting | Value | |---|---| | GPU | RTX A6000 48 GB VRAM, ~$0.49/hr | | Template | most recent RunPod PyTorch with CUDA 12.x | | Container Disk | 50 GB default | | Volume Disk | 100 GB at /workspace | | Expose | SSH port 22 — default | Click Deploy On-Demand . Wait ~30 sec until status is RUNNING . On the pod's card click Connect → SSH over exposed TCP and copy the SSH command. It looks like: ssh root@<POD IP -p <POD PORT -i ~/.ssh/id ed25519 From your Mac terminal, paste the SSH command from step 1. Type yes to accept the host key on first connect. If you get Permission denied publickey : ssh-add ~/.ssh/id ed25519 …then retry. On the pod: cd /workspace && \ git clone https://github.com/shiehn/sas-sample-generator.git && \ cd /workspace/sas-sample-generator && \ ./scripts/setup.sh 2 &1 | tee /workspace/setup.log Why these paths matter and the reason this used to be slow : /workspace is a network filesystem MooseFS — fine for big sequential reads/writes model weights, generated audio but painfully slow for many-tiny-files a Python venv . The script installs the venv at /root/.venv , which is on the pod's container-local SSD, and only keeps the HuggingFace cache and outputs on /workspace . Roughly: /root/.venv ← Python venv fast SSD; ~5 min install /workspace/sas-sample-generator ← cloned repo /workspace/.cache/huggingface ← model weights downloaded once /workspace/outputs ← generated WAVs You're done when you see: setup cuda available: True setup device: NVIDIA RTX A6000 setup done. setup next: source /root/.venv/bin/activate source /root/.venv/bin/activate hf auth login Paste your HF token One-Time Setup A . Answer n to "Add token as git credential". The 14 prompt files are already in prompts/<category .txt . To run them as-is, skip to step 6 . To customize: Edit content : nano prompts/kick.txt or scp over your own version, or edit on Mac → git push → git pull on the pod . Subset which categories run : editand comment out the lines you want to skip. Useful for prompt iteration on a single category. scripts/categories.txt Wrap the run in tmux first so an SSH drop doesn't kill the job: tmux new -s sas ./scripts/run all.sh 2 &1 | tee /workspace/run.log Detach with Ctrl-b d ; reattach later with tmux attach -t sas . The wrapper runs, in order: Build JSONLs — list to jsonl.py turns each prompts/<cat .txt into prompts/<cat .jsonl , stamping a per-category variants count from 5–6 candidates per prompt; more for high-value kick/snare/clap/808 . scripts/drum gate config.py Generate + gate, with retry-to-target — run retry.py drives batch generate.py one model load, batched — BATCH SIZE generations per call then gate drums.py , which best-of-N selects one winner per prompt and rejects clipped / silent / off-band / multi-hit / wrong-decay samples per-category profiles in drum gate config.py . Any prompt whose candidates all fail is re-rolled with fresh seeds up to MAX RETRIES=2 ; if a category still has fewer than TARGET=150 survivors, more prompt lines are topped up. Winners land in gated drums/<cat / . Post-process — postprocess oneshots.py trims, LUFS-normalizes to -16 LUFS -1 dBFS peak ceiling , and mono-downmixes the gated winners into processed/<cat / . Each WAV ships with a sibling <id .txt holding its exact prompt; the prompt is also embedded in the WAV's RIFF INFO comment chunk so Logic / Ableton / Audacity / ffprobe / macOS Get-Info show it. Merging runs is a single rsync — no manifest to reconcile. First call downloads Stable Audio 3 ~5–8 GB for medium, ~3 min, one-time . SA3 converges in ~8 diffusion steps vs 120 for SAO 1.0 , so generation is fast — with the gate + retry, the bottleneck is now the CPU gate , not generation. LUFS normalization keeps kicks, hats, and splashes at the same perceived volume in your sampler. Revert to peak-only with --normalize peak , or override with --target-lufs -14 streaming-hot / -10 commercial-hot . Useful env knobs all optional : ONLY="kick clap" LIMIT=10 ./scripts/run all.sh tiny test slice TARGET auto-0 GATE=0 ./scripts/run all.sh legacy: keep all raw, no gate BATCH SIZE=32 ./scripts/run all.sh bigger batches on an 80GB GPU TARGET=200 MAX RETRIES=3 ./scripts/run all.sh chase a higher survivor count STEPS=4 ./scripts/run all.sh cheaper/faster iteration Single-category iteration tuning prompts : Edit prompts/kick.txt, then run just that category. --skip-existing means only new/changed prompt lines hit the GPU. ONLY=kick ./scripts/run all.sh ls /workspace/outputs/processed/ Should show one subdir per enabled drum role 24 by default find /workspace/outputs/processed -name " .wav" | wc -l ~150 per category survive the gate TARGET → ~3,600 for all 24 Every <id .wav ships with a sibling <id .txt containing its generation prompt. Spot-check the pairing: test "$ find /workspace/outputs/processed -name ' .wav' | wc -l " \ = "$ find /workspace/outputs/processed -name ' .txt' | wc -l " \ && echo "wav/txt pairing OK" On the pod: cd /workspace tar czf run.tar.gz outputs/processed ls -lh run.tar.gz We use tar rather than zip because the stock RunPod PyTorch image doesn't ship zip . tar is preinstalled everywhere. tar also recurses into the per-category subdirs automatically. In a second Mac terminal don't close the SSH session yet — you still need it for step 10 : cd ~/Downloads scp -P <POD PORT root@<POD IP :/workspace/run.tar.gz . tar xzf run.tar.gz open outputs/processed Finder + QuickLook to audition <POD PORT and <POD IP are the same ones from your step-1 SSH command. The unpacked structure is one folder per category: outputs/processed/ kick/ kick-c1da23da.wav kick-e5d95885.wav ... snare-standard/ snare-standard-...wav hat-closed/ ... ...etc This is the step you will forget. The pod bills $0.49/hr for as long as it exists, whether you're using it or not. Idle overnight ≈ $12 Forgotten for a week ≈ $80 Forgotten for a month ≈ $350 In the RunPod console https://www.runpod.io/console/pods , click your pod's card → Terminate . Confirm. Termination wipes /workspace . That's fine — you have the zip on your Mac. Next month, you start fresh from step 1. sas-sample-generator/ ├── README.md ← you are here ├── stable audio open batch oneshot guide.md ← long-form background ├── requirements.txt ├── docs/SAMPLE GEN V3 PLAN.md ← v3 design rationale & locked decisions ├── prompts/ │ ├── <drum-cat .txt 24 files, 200 prompts each kick.txt, clap.txt, 808.txt, … │ └── pitched/<cat .txt 28 files, 200 prompts each pianos.txt, basses.txt, pads.txt, … ├── scripts/ │ ├── setup.sh ← bootstrap a pod venv + deps + SA3 tools │ ├── gen prompts.py ← re generate the 200-prompt corpora │ │ │ ├── run all.sh ← DRUM pipeline driver │ ├── categories.txt ← which drum roles to run │ ├── category config.py ← per-role negatives + durations │ ├── drum gate config.py ← per-role gate profiles + variant counts │ ├── gate drums.py ← drum quality gate + best-of-N │ ├── list to jsonl.py ← .txt → .jsonl drums │ ├── postprocess oneshots.py ← trim / LUFS / mono / tag │ │ │ ├── run pitched.sh ← PITCHED pipeline driver │ ├── pitched categories.txt ← which instruments to run │ ├── pitched category config.py ← target pitches, durations, zones, variants │ ├── gate pitched.py ← pitch + quality gate + best-of-N │ ├── list to jsonl pitched.py ← .txt → .jsonl multi-source fan-out │ ├── enrich pitched.py ← pitch-correct + multi-zone render + manifest │ ├── pitch report.py ← measured-vs-target pitch-accuracy report │ ├── repair instrument pitch.py ← post-hoc single-source pitch fix │ │ │ ├── run retry.py ← generate+gate+retry-to-target both pipelines │ ├── batch generate.py ← batched SA3 inference │ ├── build pack.py ← deterministic versioned pack zips │ ├── README-PACKS.md ← pack build + publish runbook │ └── benchmark.py / sync.sh ← optional helpers ├── tests/ ← CPU-only suite see "Tests" below └── outputs/ ← gitignored; all generated audio lands here ├── raw/<cat /<id vNN.wav ← SA3 candidates N variants/prompt ├── raw/<cat / metadata/<id vNN.json ← seed, model, gen params stays on pod ├── gated drums/<cat /<id .wav ← drum gate winners ├── processed/<cat /<id .{wav,txt} ← final drums trimmed/LUFS/mono; ICMT = prompt ├── gated/<cat /<id .{wav,gate.json} ← pitched gate winners + scores ├── reports/pitch summary.{json,md} ← pitch-accuracy report report stage └── instruments/<cat /<id / ← final instruments sources/, zones/, manifest.json Sibling pipeline to the drum one. Same Stable Audio 3 generator, different downstream: a 5-stage quality+pitch gate prefilter, onset, sustain-plateau, CREPE/spectral pitch with a sub-bass octave cross-check, BasicPitch polyphony when the TF/numpy ABI is happy , multi-source pitch correction , and zone pre-rendering RubberBand R3, formant-preserving . Emits a per-instrument manifest.json consumed by sas-instrument-plugin . Multi-source real-pitch sampling the v3 headline . Wide-range instruments are generated at 2–4 real source pitches spanning their natural register e.g. basses at E1/E2/E3 . Enrich assigns each playable zone to its nearest real source, so no zone is pitch-shifted more than ~half the inter-source gap — small, artifact-free shifts instead of stretching one root ±12 semitones. Toggle with SAS MULTI SOURCE=1 default; 0 reproduces v1 single-source for an A/B . 28 categories ship, 200 prompts each 5,600 prompts : | Category | Source pitch es | Dur | Var | Notes | |---|---|---|---|---| synths | C3 | 5.0s | 6 | analog mono, FM, wavetable, acid | lead-supersaw | C4·C5 | 5.0s | 6 | multi-source | lead-fm | C4 | 5.0s | 6 | | lead-acid | C3 | 4.0s | 6 | 303-style | pluck-synth | C4 | 3.0s | 6 | | plucks | C4 | 3.0s | 6 | | keys | C3 | 5.0s | 6 | Rhodes, Wurli, clav, DX7 | pianos | C2·C3·C4·C5 | 5.0s | 6 | multi-source, step 2 | organs | C3 | 8.0s | 8 | open-ended | basses | E1·E2·E3 | 6.0s | 8 | multi-source, 30 Hz floor | 808-bass | C2·C3 | 6.0s | 8 | multi-source, 25 Hz floor | reese-bass | E2·E3 | 6.0s | 8 | multi-source | pads | C3 | 12.0s | 8 | open-ended, step 2 | strings | A2·A3·A4 | 8.0s | 8 | multi-source, open-ended, step 2 | brass | A2·A3·A4 | 6.0s | 8 | multi-source, open-ended | winds | D3·D4·D5 | 5.0s | 8 | multi-source, open-ended, step 2 | accordion | F 3 | 6.0s | 8 | open-ended | bells | C5 | 4.0s | 6 | glockenspiel, FM, music box | mallets | C4 | 3.0s | 6 | marimba, vibes, kalimba | percussion | C4 | 2.0s | 6 | tonal/tuned only | timpani | F2·F3 | 4.0s | 8 | multi-source, tuned drum | guitars | E2·E3·E4 | 4.0s | 6 | multi-source | banjos | C4·G4 | 3.0s | 6 | multi-source | mandolin | A4 | 3.0s | 6 | | harp | C3·C4·C5 | 4.0s | 6 | multi-source | sitar | C4 | 4.0s | 6 | | vocals | A3 | 5.0s | 20 | choir, chops, vocoded SA3 vocals are weak | choir | A3 | 10.0s | 18 | open-ended | "Var" = SA3 candidates per prompt × source pitch ; the gate keeps the best. Pitches use the C4 = MIDI 60 convention; multi-source counts apply with SAS MULTI SOURCE=1 default . Full config in scripts/pitched category config.py /shiehn/sas-sample-generator/blob/main/scripts/pitched category config.py . To subset which categories run, edit scripts/pitched categories.txt /shiehn/sas-sample-generator/blob/main/scripts/pitched categories.txt — comment any line with to skip that category. scripts/setup.sh installs everything the pitched pipeline needs: - All Python deps from requirements.txt librosa, torchcrepe, basic-pitch, pyloudnorm, soxr, … - All system packages — rsync transfer to Mac , tmux long sessions survive disconnect , rubberband-cli enrich shells out to it directly for formant-preserving pitch shift , ffmpeg audio inspection , zip / unzip archives - The compiled stable-audio-tools from git main the PyPI release doesn't support SA3-medium - CUDA-12.8 torch wheels Blackwell-compatible, also works on older Hopper/Ampere/Ada You should never need to apt-get or pip install anything on a pod after running setup.sh . If you do, treat it as a bug in setup.sh and add it there. brew install rubberband pitch-shift backend cd ~/path/to/sas-sample-generator your local repo pip install -r requirements.txt one-time, ~5 min Designed to be safely repeatable from a cold start. The whole pipeline: ~15 min setup + ~30 min generate + ~10 min gate + ~30 min enrich + transfer . runpod.io/console/pods https://www.runpod.io/console/pods → Deploy → GPU Pod : | Setting | Value | Why | |---|---|---| | GPU | RTX A6000 / 4090 / 5090 / L40S / A100 24+ GB VRAM | SA3-medium fits in 16 GB; 24 GB gives headroom | | Template | most recent RunPod PyTorch with CUDA 12.x | matches our cu128 wheels | Container Disk | 100 GB | persistent across pod restart; holds venv + HF model cache + outputs | Network Volume | None | RunPod's "migrate to new host" flow has been known to attach a tiny 10 GB network volume — don't let it. We use container disk only | | Expose | SSH port 22, default | Critical: the field is named "Container Disk" — the persistent SSD. Do NOT confuse with "Network Volume" or "Volume Disk". Click Deploy On-Demand . Wait ~30 sec for status RUNNING . Copy the SSH command from Connect → SSH over exposed TCP . It looks like: ssh root@<POD IP -p <POD PORT -i ~/.ssh/id ed25519 ssh root@<POD IP -p <POD PORT -i ~/.ssh/id ed25519 Type yes on first connect. If Permission denied : ssh-add ~/.ssh/id ed25519 . cd /workspace && \ git clone https://github.com/shiehn/sas-sample-generator.git && \ cd /workspace/sas-sample-generator && \ ./scripts/setup.sh 2 &1 | tee /root/setup.log Look for these "OK" markers near the end of setup.log: setup rsync: rsync version 3.x.x ... setup tmux: tmux 3.x setup rubberband: /usr/bin/rubberband setup ffmpeg: ffmpeg version ... setup cuda available: True setup device: NVIDIA RTX 4090 setup done. If cuda available: False → you deployed onto a CPU template; terminate, redeploy with PyTorch GPU. source /root/.venv/bin/activate hf auth login Paste your HF read token. Answer n to "Add token as git credential". First time on a new HF account: in your browser, visit stabilityai/stable-audio-3-medium https://huggingface.co/stabilityai/stable-audio-3-medium and accept BOTH the SA3 community license AND the underlying Gemma terms. Without both, the model download fails with GatedRepoError . Token IS your account — accept while logged into the same HF account your token belongs to. Verify access under 5 seconds : hf download stabilityai/stable-audio-3-medium model config.json --local-dir /tmp/sa3-test ls /tmp/sa3-test/ If model config.json is listed: cleared. The default ships with all 16 categories enabled. For a quick test or a focused run: Edit scripts/pitched categories.txt — comment out any category with nano scripts/pitched categories.txt After the test, restore with git checkout scripts/pitched categories.txt . Run generate + gate + report on the GPU pod; enrich runs later on your Mac it's CPU-bound . tmux new -s pitched Inside tmux: cd /workspace/sas-sample-generator source /root/.venv/bin/activate source /workspace/.bash env STAGES=generate,gate,report ./scripts/run pitched.sh 2 &1 | tee outputs/run.log run pitched.sh builds the JSONLs multi-source fan-out — one job per prompt × source pitch, with the per-category variants count , then run retry.py drives batched generation + gate pitched.py with the same retry-to-target loop as drums re-roll all-fail prompts up to MAX RETRIES=2 , top up until TARGET=150 instruments survive . The report stage then writes outputs/ reports/pitch summary.{json,md} so you can read measured-vs-target pitch accuracy before transferring. Detach with Ctrl-b d. The run keeps going even if SSH drops. Reattach later from any new SSH session — possibly a new IP/port if migrated : tmux attach -t pitched Monitor from outside tmux: tail -f outputs/run.log nvidia-smi Throughput: batched generation is ~1–3 h; the CPU gate is the bottleneck torchcrepe + basic-pitch + librosa per variant . For all 28 categories at full variant counts, budget up to ~a day of pod time. Dial variant counts down in pitched category config.py , or run a subset, if wall-clock matters. Useful knobs: ONLY=pianos LIMIT=5 tiny slice , BATCH SIZE=32 80 GB GPU , MAX RETRIES=0 one pass, no retry , INIT ANCHOR=1 experimental init audio pitch anchoring — default off , SAS MULTI SOURCE=0 single-source A/B . When STAGES=generate,gate finishes, before transferring: Per-category pass rates for d in outputs/gated/ /; do cat=$ basename "$d" "$cat" == " failures" && continue passed=$ ls "$d" .wav 2 /dev/null | wc -l failed=$ ls "$d/ failures"/ .json 2 /dev/null | wc -l total=$ passed + failed if $total -gt 0 ; then rate=$ passed 100 / total printf " %-18s passed=%3d failed=%3d pass-rate=%d%%\n" "$cat" "$passed" "$failed" "$rate" fi done echo "Total gated: $ find outputs/gated -name ' .wav' -not -path ' failures ' | wc -l " du -sh outputs/gated Expected with the current thresholds, 2026-05-22 : 80–100% pass rate per category . If a category is below 50%, look in outputs/gated/<cat / failures/<id .json to see why prompts are failing. The pod has rsync installed by setup.sh . On your Mac: mkdir -p ~/sas-pitched-out rsync -avzP -e "ssh -p <POD PORT -i ~/.ssh/id ed25519" \ root@<POD IP :/workspace/sas-sample-generator/outputs/gated/ \ ~/sas-pitched-out/gated/ For ~4 GB at typical RunPod / home upload speeds, expect 10–20 min. rsync resumes on interruption — just re-run the same command if SSH drops. Verify locally: find ~/sas-pitched-out/gated -name ' .wav' -not -path ' failures ' | wc -l should match step 7 du -sh ~/sas-pitched-out/gated cd ~/path/to/sas-sample-generator git pull pick up any threshold updates pip install -r requirements.txt idempotent export SAS OUTPUTS DIR=~/sas-pitched-out STAGES=enrich ./scripts/run pitched.sh Enrich groups the surviving source pitches of each prompt into one multi-source instrument under ~/sas-pitched-out/instruments/<cat /<id / : sources/<midi .wav — 24-bit real source samples 1 per source pitch: 2–4 for multi-source categories , pitch-corrected + normalized to -20 LUFS zones/<midi .wav — 16-bit WAV pre-rendered playable zones every 2–3 semitones , each rendered from its nearest real source was 24-bit FLAC pre-v3; WAV is memory-mapped by the Tracktion sampler with no decode stall manifest.json — schema version: 1 , disjoint ordered zones prompt.txt — original positive prompt It parallelizes across instruments ProcessPoolExecutor and shells out to the rubberband CLI for pitch shifts brew install rubberband on the Mac . runpod.io/console/pods https://www.runpod.io/console/pods → pod card → Terminate NOT Stop . Compute billing stops immediately. Volume billing if any auto-created Network Volume snuck in stops only on Terminate. Then runpod.io/console/user/storage https://www.runpod.io/console/user/storage → Network Volumes → check for any outside orphan from a migration → Delete. RunPod sometimes moves your pod to a different physical host mid-run. Symptoms: - SSH connection drops mid-session Connection refused when reconnecting on the same IP/port- Pod shows "Stopped" briefly, then "Running" again at a new address The pod, the venv, the HF cache, and all outputs/ data persist on the container disk as long as the pod isn't terminated. You just need fresh connection info. - Open RunPod console → click your pod card → check the Connect → SSH over exposed TCP panel for the new IP and port both can change . - Clear the old SSH host key on your Mac: ssh-keygen -R ' <NEW IP :<NEW PORT ' - SSH back in with the new details. Run tmux attach -t pitched — your run is still going. - If you were mid-rsync, just re-run the rsync command with the new -p <NEW PORT and root@<NEW IP — it picks up where it stopped. This bit us twice this session May 2026 . Symptoms are unambiguous; recovery takes 30 seconds. SA3 doesn't reliably hit a target pitch from a text prompt — that's a known limitation of text-to-audio diffusion models. Enrich now compensates intelligently: | If measured pitch is… | Enrich does… | Result | |---|---|---| within max correction semitones of target default 3 | shifts all the way to the original target | Sample is at exactly the prompted MIDI note; preserves prompt semantics | | further away than that | snaps to the nearest integer semitone | Sample is at the closest "logical" MIDI note always ≤50 cent shift, no audible artifacts | Either way: every output sample lands on an exact MIDI semitone with the smallest possible pitch shift. The zone rendering loop centers on that effective root, so the sampler always has a clean zone at the sample's actual pitch. max correction semitones is per-category in scripts/pitched category config.py . Set to 0 to always snap to nearest semitone never shift to target . Set to a large value 24+ to always shift to target. | Stage | What it checks | What rejection means | |---|---|---| prefilter | Clipping, dead channels, all-silent buffers | Sample is broken at the file level | onset | Time from buffer start to first transient | slow onset → SA3 added a fade-in / silence preamble 300ms | sustain | Longest plateau within 12 dB of peak RMS | short stab → audio decays too fast or has no held region | pitch | CREPE periodicity + measured-vs-target | no voiced frames / unconfident → unpitched output; tolerance 9999 so enrich's snap-to-nearest-semitone can do its job wrong pitch is OFF by default | polyphony | BasicPitch note count after vibrato bypass | Disabled when TF/numpy ABI mismatches common on RunPod — gate prints one warning at start, then runs without it | The gate scores winners by confidence² × exp -|cents|/50 × sus quality . With wrong pitch disabled, the pitch term collapses to ~0 for far-off samples, so all variants of a prompt can tie at score=0.000 — the picker just grabs v00 by default in that case. Acceptable for now. outputs/ ├── raw/<category / ← SA3 candidates N variants per prompt × source pitch │ ├── <id v00.wav, <id v01.wav, ... │ └── metadata/<id v0N.json ← seed, model, generation seconds, batch size ├── gated/<category / ← gate winners only │ ├── <id .wav ← chosen variant │ ├── <id .gate.json ← per-gate scores + measured pitch │ └── failures/<id .json ← prompts where ALL variants rejected ├── reports/pitch summary.{json,md} ← measured-vs-target accuracy report stage └── instruments/<category /<instrument-id / ← final library, sampler-consumable ├── sources/<midi .wav ← 24-bit real source pitches 1–4 ├── zones/<midi .wav ← 16-bit WAV pre-rendered zones ├── manifest.json ← schema version 1, disjoint ordered zones └── prompt.txt ← original positive prompt The v3 full run is much bigger than a quick slice: 28 categories × 200 prompts × 2–4 source pitches × 6–20 variants = tens of thousands of candidates. Generation is cheap and batched; the CPU gate dominates wall-clock per-variant CREPE + basic-pitch + librosa . On an A100 80 GB ~~$0.89–1.89/hr the whole pitched campaign runs in roughly ~~$15–35 , with drums adding a few dollars. The original $200 budget is never the binding constraint. half a day to a day → To cut cost/time: run fewer categories ONLY=… , lower variant counts in pitched category config.py , or set MAX RETRIES=0 . A single-category slice ONLY=pianos LIMIT=5 is a few minutes and a few cents — use it to dial prompts before committing the full campaign. Enrich is local on your Mac → $0 . prompts/pitched/<category .txt one prompt per line, comments scripts/pitched categories.txt which categories to run comment to skip scripts/pitched category config.py per-category target pitch, duration, sustain thresholds, etc. Fast iteration on a single category: 1. ONLY=<cat overrides the enable-list — no need to edit pitched categories.txt 2. Edit prompts/pitched/<cat .txt 3. Re-run generate+gate+report LIMIT caps prompts; sets TARGET=0 ONLY=<cat LIMIT=10 STAGES=generate,gate,report ./scripts/run pitched.sh 4. Listen to outputs/gated/<cat / .wav, read outputs/ reports/pitch summary.md, re-run Re-author / extend the corpora with scripts/gen prompts.py /shiehn/sas-sample-generator/blob/main/scripts/gen prompts.py deterministic, preserves existing lines so hashes stay stable, holds the EDM ratio . --skip-existing in batch generate.py means re-running won't regenerate samples you already have — only new prompt lines hit the GPU. STAGES=generate,gate,enrich,report — pitched only; comma-separated subset default: all four STEPS=8 — diffusion steps SA3 converges fast BATCH SIZE=16 — generations per model call 32–64 on an 80 GB GPU TARGET=150 — per-category minimum surviving samples 0 = no top-up MAX RETRIES=2 — re-roll rounds for all-fail prompts before topping up 0 = one pass ONLY="a b" / LIMIT=N — run a subset of categories / cap prompts per category test slice; auto-sets TARGET=0 GATE=0 — drums only; skip the quality gate legacy keep-all INIT ANCHOR=1 — pitched only; experimental init audio pitch anchoring default off SAS MULTI SOURCE=0 — pitched only; disable multi-source single root, span 12; for an A/B SAS OUTPUTS DIR=/path — override outputs location default /workspace/outputs on pod, ./outputs local Per-prompt variant counts are per-category now in drum gate config.py / pitched category config.py , not a global env var. The sas-instrument-plugin walks outputs/instruments/<cat /<id / , parses each manifest.json , and uses the zones array to call host.setTrackInstrumentSampler on the chosen track. Disjoint zones + per-zone root midi mean the engine pitch-shifts the nearest pre-rendered zone for any played MIDI note, with the smart-corrected sample as the unshifted root. Since enrich locks every sample to an integer MIDI semitone, the sampler never has to deal with off-pitch sources. Generated audio ships to the Signals & Sorcery https://signalsandsorcery.com app as versioned pack zips built by scripts/build pack.py /shiehn/sas-sample-generator/blob/main/scripts/build pack.py — deterministic fixed mtimes + sorted entries → byte-identical zip + sha256 from an identical source tree : | Pack | Source dir | Zip | Approx size v3 | |---|---|---|---| | Drums | outputs/processed/ | sas-drum-pack-v{N}.zip | ~2–3 GB 24 roles | | Instruments | outputs/instruments/ | sas-instrument-pack-v{N}.zip | ~20–24 GB 28 cats × ~150 | | Loops | outputs/loops/ | sas-loop-library-v{N}.zip | external loop library not generated here | python scripts/build pack.py --smoke-test cheap determinism check python scripts/build pack.py --pack drums --version 1 real build → ./dist/ python scripts/build pack.py --pack instruments --version 1 Ready-to-consume directories no zip / download step . To produce the two libraries as folders you can drop straight into the app's install location — <userData /sample-packs/{drums,instruments}/ — use --format dir , or the wrapper. Each emits dist/<subdir / with the pack-version.json marker at its root: exactly the tree the app expects on disk, ready out of the box. after gating + enriching processed/ + instruments/ are populated : DRUM VERSION=3 INSTRUMENT VERSION=3 ./scripts/build libraries.sh → dist/drums/, dist/instruments/ drop into your local app macOS path shown : rsync -a dist/drums/ ~/Library/Application\ Support/signals-and-sorcery/sample-packs/drums/ rsync -a dist/instruments/ ~/Library/Application\ Support/signals-and-sorcery/sample-packs/instruments/ The marker version must equal that pack's expectedVersion in sas-app/src/shared/constants/sample-packs.ts plain string match or the app treats the folder as a different version. For GCP distribution use --format zip default and follow the publish runbook below. The build prints sizeBytes + sha256 ; paste those into sas-app/src/shared/constants/sample-packs.ts bump expectedVersion + the download URL so the app detects the new version and prompts a re-download. Never overwrite a published version — always bump. Full publish runbook GCP upload, version rules, what's in/excluded, the v3 WAV-zone note : scripts/README-PACKS.md /shiehn/sas-sample-generator/blob/main/scripts/README-PACKS.md . CPU-only suite — no GPU, no Stable Audio. It covers everything around model generation: pitch detection, both gates, multi-source enrich, retry-to-target helpers, config/prompt/enable wiring, loudness targets, list to jsonl , the pitch report, and the deterministic pack builder. ./tests/run tests.sh Keep it green — run before and after any pipeline-code change. Deps live in the project .venv numpy, soundfile, pyloudnorm, librosa ; the rubberband CLI is optional the enrich test degrades gracefully without it . | Symptom | Most likely cause | Fix | |---|---|---| Permission denied publickey on ssh | private key not loaded into agent | ssh-add ~/.ssh/id ed25519 | setup.sh hangs at Installing collected packages: for 5 min | something redirected the venv onto /workspace MooseFS ; script defaults to /root/.venv for a reason | check echo $VENV DIR — should be /root/.venv . If overridden, unset it and re-run | cuda available: False after setup.sh | picked a CPU template | terminate; re-deploy with PyTorch GPU template | huggingface hub.utils. errors.GatedRepoError | didn't accept the SA3 license or the Gemma terms it inherits | visit the | batch generate.py errors CUDA out of memory --default-duration or --num-waveforms-per-prompt 1 one shot, no loop, no hi hats, no snare to every prompt dry, no reverb, no ambience to prompts negative prompt may be excluding the target — bug in scripts/category config.py scripts/category config.py , audit the negative for that category; nothing in it should match the target sound run all.sh skips a category prompts/<cat .txt is missing or has only comments prompts/ has the .txt file and contains non-comment lines rejected/ with reason=lufs unmeasurable pyloudnorm and check they cluster near --target-lufs . The pipeline log at /workspace/run.log from step 6 records the as-run postprocess flags for each category. tmux new -s sas BEFORE running, reattach with tmux attach -t sas Stable Audio 3 needs only ~8 diffusion steps vs 120 for SAO 1.0 , so generation is fast and batched. Since v3 added the quality gate + retry-to-target, the CPU gate — not generation — sets the wall-clock , and the per-category minimum-survivor target TARGET=150 re-rolls until each category fills. | Run shape | Where the time goes | Rough cost | |---|---|---| Single-category slice ONLY=… LIMIT=… | a few minutes, gate-bound | a few cents | Full drum run 24 roles, gate + retry | ~1 h on a big GPU | ~$1–3 | Full pitched run 28 cats, multi-source, gate + retry | ½–1 day, gate-bound | ~$15–35 | Keep a pod alive between iterations and --skip-existing skips anything already generated — so re-running after a prompt tweak only pays for the new lines. Terminate as soon as generate + gate finish enrich is local and free . This is a public repo . Never commit: - Hugging Face tokens - RunPod API keys - B2 / R2 / S3 keys - SSH private keys - The generated WAVs gitignored already .gitignore covers .env , .token , .secret , outputs/ . If you ever git add a file containing a secret by mistake: rotate the secret first , then git rm + commit + push. Treat anything that hit main as compromised. stable audio open batch oneshot guide.md /shiehn/sas-sample-generator/blob/main/stable audio open batch oneshot guide.md covers: - Why Stable Audio 3 vs alternatives and what changed from SAO 1.0 - Prompt-design rules and category-specific templates - Optional persistent Network Volume layout for users running multiple times per week - Optional rclone push to Backblaze B2 / Cloudflare R2 instead of scp - Optional custom Docker image - Cost-control deep dive This is one piece of a larger ecosystem around the Signals & Sorcery https://signalsandsorcery.com audio app. Plugin SDK & templates sas-plugin-sdk https://github.com/shiehn/sas-plugin-sdk — types, components, and hooks for building generator plugins sas-plugin-template https://github.com/shiehn/sas-plugin-template — starter template for new plugins sas-chat-plugin https://github.com/shiehn/sas-chat-plugin — in-app conversational agent Built-in plugins sas-stems-plugin https://github.com/shiehn/sas-stems-plugin — default AI audio-from-text + stem-splitting plugin sas-loops-plugin https://github.com/shiehn/sas-loops-plugin — default audio loop / sample plugin sas-synth-plugin https://github.com/shiehn/sas-synth-plugin — default synth plugin sas-texture-plugin https://github.com/shiehn/sas-texture-plugin — texture/ambient plugin sas-recorder-plugin https://github.com/shiehn/sas-recorder-plugin — line-in recording plugin Audio tooling sas-audio-processor https://github.com/shiehn/sas-audio-processor — audio processing utilities Signals2Surge https://github.com/shiehn/Signals2Surge — synth patch transfer to Surge XT Infrastructure signals-and-sorcery-server https://github.com/shiehn/signals-and-sorcery-server — DAWNet API + WebSocket server signals-and-sorcery-docs https://github.com/shiehn/signals-and-sorcery-docs — public docs Other signalsandsorcery-game-ui https://github.com/shiehn/signalsandsorcery-game-ui — LLM-powered RPG frontend SignalsAndSorcery https://github.com/shiehn/SignalsAndSorcery — earlier VueJS + Web Audio sample arrangement tool Errantry https://github.com/shiehn/Errantry — E2E testing for agent-facing CLIs drives this project's CLI too