Datasets:
feat: add developer docs links (docs, reference, guides)
Browse files
README.md
CHANGED
|
@@ -171,8 +171,47 @@ row in the leaderboard:
|
|
| 171 |
|
| 172 |
In other words: **this dataset is also a benchmark of StemSplit's own
|
| 173 |
quality**. We didn't add a separate `stemsplit_api` row because it would
|
| 174 |
-
just duplicate those numbers.
|
| 175 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 176 |
|
| 177 |
---
|
| 178 |
|
|
@@ -217,14 +256,13 @@ PyTorch 2.11 with the MPS backend, for the v1 lineup:
|
|
| 217 |
|
| 218 |
We run [StemSplit](https://stemsplit.io) — a hosted stem-separation service —
|
| 219 |
and we needed an honest, public, reproducible way to compare ourselves to the
|
| 220 |
-
state of the art. So we open-sourced it.
|
| 221 |
-
product without standing up GPU infrastructure, our API is a drop-in:
|
| 222 |
|
| 223 |
-
|
| 224 |
-
|
| 225 |
-
|
| 226 |
-
|
| 227 |
-
|
| 228 |
|
| 229 |
---
|
| 230 |
|
|
|
|
| 171 |
|
| 172 |
In other words: **this dataset is also a benchmark of StemSplit's own
|
| 173 |
quality**. We didn't add a separate `stemsplit_api` row because it would
|
| 174 |
+
just duplicate those numbers.
|
| 175 |
+
|
| 176 |
+
---
|
| 177 |
+
|
| 178 |
+
## Use the StemSplit API
|
| 179 |
+
|
| 180 |
+
If you'd rather not stand up Demucs, ffmpeg, torchcodec, and a GPU yourself,
|
| 181 |
+
the StemSplit API ships the same models with a single HTTP call. Pricing,
|
| 182 |
+
quotas, and the full feature set are documented in our developer portal:
|
| 183 |
+
|
| 184 |
+
| Resource | URL |
|
| 185 |
+
|---|---|
|
| 186 |
+
| 🏠 Developer landing | [stemsplit.io/developers](https://stemsplit.io/developers) |
|
| 187 |
+
| 📘 Getting-started docs (auth, upload, polling) | [stemsplit.io/developers/docs](https://stemsplit.io/developers/docs) |
|
| 188 |
+
| 📑 API reference (every endpoint, every field) | [stemsplit.io/developers/reference](https://stemsplit.io/developers/reference) |
|
| 189 |
+
| 🧩 Integration guides (Zapier, n8n, Make, Pipedream, Discord, Audacity, DJ workflows, ...) | [stemsplit.io/developers/guides](https://stemsplit.io/developers/guides) |
|
| 190 |
+
|
| 191 |
+
Minimal example — submit a job, poll, download stems:
|
| 192 |
+
|
| 193 |
+
```bash
|
| 194 |
+
# 1. Submit
|
| 195 |
+
JOB=$(curl -sS -X POST https://stemsplit.io/api/v1/jobs \
|
| 196 |
+
-H "Authorization: Bearer $STEMSPLIT_API_KEY" \
|
| 197 |
+
-F "file=@song.wav" \
|
| 198 |
+
-F "stems=4")
|
| 199 |
+
JOB_ID=$(echo "$JOB" | jq -r .id)
|
| 200 |
+
|
| 201 |
+
# 2. Poll until completed
|
| 202 |
+
while [ "$(curl -sS https://stemsplit.io/api/v1/jobs/$JOB_ID \
|
| 203 |
+
-H "Authorization: Bearer $STEMSPLIT_API_KEY" | jq -r .status)" != "completed" ]; do
|
| 204 |
+
sleep 3
|
| 205 |
+
done
|
| 206 |
+
|
| 207 |
+
# 3. Download every stem
|
| 208 |
+
curl -sS https://stemsplit.io/api/v1/jobs/$JOB_ID \
|
| 209 |
+
-H "Authorization: Bearer $STEMSPLIT_API_KEY" \
|
| 210 |
+
| jq -r '.stems | to_entries[] | "\(.key) \(.value)"' \
|
| 211 |
+
| while read stem url; do curl -sSL "$url" -o "$stem.wav"; done
|
| 212 |
+
```
|
| 213 |
+
|
| 214 |
+
→ Get an API key at [stemsplit.io/developers](https://stemsplit.io/developers).
|
| 215 |
|
| 216 |
---
|
| 217 |
|
|
|
|
| 256 |
|
| 257 |
We run [StemSplit](https://stemsplit.io) — a hosted stem-separation service —
|
| 258 |
and we needed an honest, public, reproducible way to compare ourselves to the
|
| 259 |
+
state of the art. So we open-sourced it.
|
|
|
|
| 260 |
|
| 261 |
+
Want to ship a separation product without standing up GPU infrastructure?
|
| 262 |
+
The same models are one HTTP call away — see the
|
| 263 |
+
[Use the StemSplit API](#use-the-stemsplit-api) section above, or jump
|
| 264 |
+
straight to the [developer docs](https://stemsplit.io/developers/docs)
|
| 265 |
+
and [API reference](https://stemsplit.io/developers/reference).
|
| 266 |
|
| 267 |
---
|
| 268 |
|