feat: OGG/Opus import and OGG export (#330)

README.md

@@ -28,9 +28,9 @@
 
 <br>
 
-Drop in an MP3, WAV, or FLAC file, or paste a YouTube URL, and StemDeck splits the audio into up to six stems (vocals, drums, bass, guitar, piano, other). Play them back in a DAW-style multitrack mixer: mute, solo, balance levels, zoom the waveform, loop a region, and export individual stems or a custom mix. Everything runs locally on your own machine.
+Drop in an MP3, WAV, FLAC, OGG/Opus, MP4, or M4A file, or paste a YouTube URL, and StemDeck splits the audio into up to six stems (vocals, drums, bass, guitar, piano, other). Play them back in a DAW-style multitrack mixer: mute, solo, balance levels, zoom the waveform, loop a region, and export individual stems or a custom mix. Everything runs locally on your own machine.
 
-> **What is this?** StemDeck is a stem separation tool, not a downloader. Its main job is processing audio you already own: drag an MP3, WAV, or FLAC onto the import bar and go. YouTube support is a convenience for content you have the right to process. StemDeck does not store, cache, or redistribute any downloaded content. Everything happens locally and nothing leaves your machine.
+> **What is this?** StemDeck is a stem separation tool, not a downloader. Its main job is processing audio you already own: drag an MP3, WAV, FLAC, OGG, or M4A onto the import bar and go. YouTube support is a convenience for content you have the right to process. StemDeck does not store, cache, or redistribute any downloaded content. Everything happens locally and nothing leaves your machine.
 
 > StemDeck is a free, open alternative to cloud stem-splitters like Moises and LALAL.AI: no account, no quota, no uploads, no subscription. If you want stems for personal study and prefer to keep things local and free, StemDeck has you covered. If you need the polish, a mobile app, or deeper musician tooling, the commercial products are a better fit.
 
@@ -58,7 +58,7 @@ StemDeck is free and **does not accept any money, sponsorship, or funding** - no
 
 **6-stem separation** via Demucs `htdemucs_6s`, with auto-detection of the best Torch device (CUDA on NVIDIA, MPS on Apple Silicon, CPU fallback).
 
-**YouTube and local file import.** Paste a YouTube URL or drop an MP3 or WAV directly onto the import bar.
+**YouTube and local file import.** Paste a YouTube URL or drop an MP3, WAV, FLAC, OGG/Opus, MP4, or M4A directly onto the import bar.
 
 **DAW-style waveform editor** with min/max sample rendering across all stems, shared normalization, zoom in/out/Fit, loop drag on the ruler, gold playhead overlay, and stem-aligned lanes.
 
@@ -94,7 +94,7 @@ StemDeck is not trying to compete with commercial stem-separation products. It c
 | **Data retention** | You control it; delete anytime | Governed by their privacy policy and retention period |
 | **Stem model** | Demucs `htdemucs_6s` (open source, Meta AI) | Proprietary models, regularly updated, generally higher quality |
 | **Stem count** | 6 (vocals, drums, bass, guitar, piano, other) | Up to 10 depending on service and plan |
-| **Input formats** | YouTube URL, MP3, WAV | MP3, WAV, FLAC, M4A, and more depending on service |
+| **Input formats** | YouTube URL, MP3, WAV, FLAC, OGG/Opus, MP4, M4A | MP3, WAV, FLAC, M4A, and more depending on service |
 | **Processing speed** | Depends on your hardware; fast with a GPU, slow on CPU only | Fast regardless of your hardware (runs on their servers) |
 | **Batch processing** | One job at a time | Yes, on paid plans |
 | **Mobile app** | No | iOS and Android |
@@ -284,7 +284,7 @@ The library is persistent by default (`STEMDECK_PERSIST_LIBRARY=1`), so tracks a
 ## How to Use
 
 1. On the import bar, click stem chips to choose which stems to extract (defaults to all 6).
-2. Paste a YouTube URL **or** drop an MP3/WAV file, then click **Process**.
+2. Paste a YouTube URL **or** drop an audio file (MP3, WAV, FLAC, OGG, MP4, M4A), then click **Process**.
 3. Wait through `Uploading...` / `Downloading...` → `Analyzing...` → `Separating...` → `Mixing tracks...`.
 4. When done, the studio dashboard appears. If you picked a subset, the first lane is **Original** (full song minus your selection); the rest are your isolated stems.
 5. Mix: **Play/Pause/Stop** controls the master transport. **M** mutes a stem, **S** solos it (additive; multiple solos stay audible), **Monitor** solos only that stem and clears others. The volume fader moves 1:1 with drag; double-click resets to 0 dB; `Shift+wheel` gives coarse adjustment and plain wheel gives fine. The **Reset**, **Mute**, and **Solo** toolbar buttons act on all stems at once.

app/api/jobs.py

@@ -27,7 +27,7 @@
 router = APIRouter(tags=["jobs"])
 logger = logging.getLogger("stemdeck.api")
 
-_ALLOWED_EXTS = frozenset((".mp3", ".wav", ".flac", ".mp4", ".m4a"))
+_ALLOWED_EXTS = frozenset((".mp3", ".wav", ".flac", ".mp4", ".m4a", ".ogg", ".opus"))
 _MAX_UPLOAD_BYTES = 400 * 1024 * 1024  # 400 MB
 _WS_RE = re.compile(r"\s+")
 
@@ -171,7 +171,7 @@ async def _create_local_job(request: Request) -> dict[str, str]:
     if ext not in _ALLOWED_EXTS:
         raise HTTPException(
             status_code=422,
-            detail=f"Unsupported file type '{ext}': accepted formats are .mp3, .wav, .flac, .mp4, and .m4a",
+            detail=f"Unsupported file type '{ext}': accepted formats are .mp3, .wav, .flac, .mp4, .m4a, .ogg, and .opus",
         )
 
     # Validate stems list from form field

app/api/stems.py

@@ -45,14 +45,21 @@
 _MIXDOWN_MAX_GAIN = 4.0
 
 # Output encoders by container/extension, shared by the dynamic mixdown and the
-# stems zip. WAV is lossless PCM, FLAC is lossless compressed, MP3 is VBR ~190 kbps.
+# stems zip. WAV is lossless PCM, FLAC is lossless compressed, MP3 is VBR ~190 kbps,
+# OGG is Vorbis VBR q6 (~192 kbps) — the quality tier matching the MP3 setting.
 _ENCODE_ARGS = {
     "wav": ["-c:a", "pcm_s16le"],
     "mp3": ["-q:a", "2"],
     "flac": ["-c:a", "flac"],
+    "ogg": ["-c:a", "libvorbis", "-q:a", "6"],
 }
 MIXDOWN_CODECS = {ext: [*args, "-f", ext] for ext, args in _ENCODE_ARGS.items()}
-MIXDOWN_MEDIA_TYPES = {"wav": "audio/wav", "mp3": "audio/mpeg", "flac": "audio/flac"}
+MIXDOWN_MEDIA_TYPES = {
+    "wav": "audio/wav",
+    "mp3": "audio/mpeg",
+    "flac": "audio/flac",
+    "ogg": "audio/ogg",
+}
 
 # Mixdown render cache (#290): identical render params re-run the full ffmpeg
 # graph on every request today. On a shared server, repeat downloads of the
@@ -403,15 +410,15 @@ async def get_mixdown(
     end: float | None = Query(default=None, gt=0, description="Trim end in seconds"),
 ) -> FileResponse | StreamingResponse:
     """Render a mixdown of the given lanes at the given gains, streamed as WAV,
-    MP3, or FLAC. Mirrors the studio mixer (per-stem volume, mute, solo) so the
+    MP3, FLAC, or OGG. Mirrors the studio mixer (per-stem volume, mute, solo) so the
     exported file matches what is heard. The master fader is intentionally not
     applied -- it is a monitoring level, not part of the mix. Optional ?start=&end=
     trims to a loop region.
 
     Identical params (including start/end and the current export sample rate)
     hit a render cache instead of re-running ffmpeg (#290) -- a cheap win on a
     shared server where the same export gets re-downloaded."""
-    if ext not in ("wav", "mp3", "flac"):
+    if ext not in ("wav", "mp3", "flac", "ogg"):
         raise HTTPException(status_code=404, detail="not found")
 
     names, parsed_gains = _parse_lane_gains(stems, gains)
@@ -559,8 +566,8 @@ async def get_video_mixdown(
 
 
 def _build_stems_zip(sources: list[tuple[str, Path]], fmt: str, dest: Path) -> None:
-    """Blocking: write the stems into a ZIP. WAV files are stored as-is; MP3 and
-    FLAC are transcoded per stem via ffmpeg. ZIP_STORED throughout - audio doesn't
+    """Blocking: write the stems into a ZIP. WAV files are stored as-is; MP3,
+    FLAC, and OGG are transcoded per stem via ffmpeg. ZIP_STORED throughout - audio doesn't
     meaningfully compress, and STORED keeps the build fast. Runs in a thread."""
     if fmt == "wav":
         with zipfile.ZipFile(dest, "w", zipfile.ZIP_STORED) as zf:
@@ -608,8 +615,8 @@ async def get_all_stems_zip(
     every available stem is included."""
     if not JOB_ID_RE.match(job_id):
         raise HTTPException(status_code=404, detail="job not found")
-    if fmt not in ("wav", "mp3", "flac"):
-        raise HTTPException(status_code=422, detail="format must be 'wav', 'mp3', or 'flac'")
+    if fmt not in ("wav", "mp3", "flac", "ogg"):
+        raise HTTPException(status_code=422, detail="format must be 'wav', 'mp3', 'flac', or 'ogg'")
     job = registry_get(job_id)
     if job is None or job.status != "done":
         raise HTTPException(status_code=404, detail="job not ready")

static/index.html

@@ -57,7 +57,7 @@
                 <path d="M12 3v12 M7 8l5-5 5 5 M5 21h14"/>
               </svg>
             </button>
-            <input id="fileInput" type="file" accept=".mp3,.wav,.flac,.mp4,.m4a,audio/mpeg,audio/wav,audio/flac,video/mp4,audio/mp4" style="display:none" aria-hidden="true" />
+            <input id="fileInput" type="file" accept=".mp3,.wav,.flac,.mp4,.m4a,.ogg,.opus,audio/mpeg,audio/wav,audio/flac,video/mp4,audio/mp4,audio/ogg,audio/opus" style="display:none" aria-hidden="true" />
           </div>
 
           <!-- Divider -->
@@ -613,6 +613,7 @@
                   <button class="export-fmt active" id="t-fmt-wav" type="button" role="radio" aria-checked="true">WAV</button>
                   <button class="export-fmt" id="t-fmt-mp3" type="button" role="radio" aria-checked="false">MP3</button>
                   <button class="export-fmt" id="t-fmt-flac" type="button" role="radio" aria-checked="false">FLAC</button>
+                  <button class="export-fmt" id="t-fmt-ogg" type="button" role="radio" aria-checked="false">OGG</button>
                   <button class="export-fmt export-fmt-video" id="t-fmt-mp4" type="button" role="radio" aria-checked="false">MP4</button>
                 </div>
                 <button class="chip-panel-item export-item" id="t-export-mix" type="button" role="menuitem">

static/js/main.js

@@ -130,6 +130,7 @@ function wireFooterControls() {
   const fmtWav   = document.getElementById("t-fmt-wav");
   const fmtMp3   = document.getElementById("t-fmt-mp3");
   const fmtFlac  = document.getElementById("t-fmt-flac");
+  const fmtOgg   = document.getElementById("t-fmt-ogg");
   const fmtMp4   = document.getElementById("t-fmt-mp4");
   const exportWrap = document.getElementById("footer-export-wrap");
   const itemMix    = document.getElementById("t-export-mix");
@@ -163,7 +164,7 @@ function wireFooterControls() {
 
   function setFormat(f) {
     format = f;
-    for (const [btn, val] of [[fmtWav, "wav"], [fmtMp3, "mp3"], [fmtFlac, "flac"], [fmtMp4, "mp4"]]) {
+    for (const [btn, val] of [[fmtWav, "wav"], [fmtMp3, "mp3"], [fmtFlac, "flac"], [fmtOgg, "ogg"], [fmtMp4, "mp4"]]) {
       btn?.classList.toggle("active", f === val);
       btn?.setAttribute("aria-checked", String(f === val));
     }
@@ -172,6 +173,7 @@ function wireFooterControls() {
   fmtWav?.addEventListener("click", (e) => { e.stopPropagation(); setFormat("wav"); });
   fmtMp3?.addEventListener("click", (e) => { e.stopPropagation(); setFormat("mp3"); });
   fmtFlac?.addEventListener("click", (e) => { e.stopPropagation(); setFormat("flac"); });
+  fmtOgg?.addEventListener("click", (e) => { e.stopPropagation(); setFormat("ogg"); });
   fmtMp4?.addEventListener("click", (e) => { e.stopPropagation(); setFormat("mp4"); });
 
   // MP4 exports the mix muxed with the source video. Stems and region have no
@@ -304,8 +306,9 @@ function wireFileDrop() {
     if (!file) return;
     const lower = file.name.toLowerCase();
     if (!lower.endsWith(".mp3") && !lower.endsWith(".wav") && !lower.endsWith(".flac") &&
-        !lower.endsWith(".mp4") && !lower.endsWith(".m4a")) {
-      showError("Only MP3, WAV, FLAC, MP4, and M4A files are supported.");
+        !lower.endsWith(".mp4") && !lower.endsWith(".m4a") &&
+        !lower.endsWith(".ogg") && !lower.endsWith(".opus")) {
+      showError("Only MP3, WAV, FLAC, MP4, M4A, OGG, and Opus files are supported.");
       return;
     }
     if (file.size > MAX_UPLOAD_BYTES) {

static/mobile/app.js

@@ -549,7 +549,7 @@ function extractScreen() {
       <div class="sub">Paste a link or upload audio to split into stems.</div>
       <div class="paste">${ICON.link}<input id="ext-url" class="ext-input" type="url" inputmode="url" autocomplete="off" autocapitalize="off" spellcheck="false" placeholder="Paste YouTube or audio URL" value="${esc(state.extractUrl || "")}"></div>
       <button class="upload" data-action="pick-file">${ICON.upload}${fileName ? esc(fileName) : "Upload audio file"}</button>
-      <input id="ext-file" type="file" accept="audio/*,video/mp4,.mp3,.wav,.flac,.m4a,.ogg" style="display:none">
+      <input id="ext-file" type="file" accept="audio/*,video/mp4,.mp3,.wav,.flac,.m4a,.ogg,.opus" style="display:none">
       <div class="eyebrow">STEMS TO EXTRACT</div>
       <div class="chips">${EXTRACT_STEMS.map((s) => {
         const on = !!state.selected[s.id];

tests/test_jobs_api.py

@@ -134,10 +134,10 @@ def test_upload_503_when_queue_full(upload_client):
 
 
 def test_upload_rejects_unsupported_extension(upload_client):
-    data = io.BytesIO(b"OGG data")
+    data = io.BytesIO(b"FORM\x00\x00\x00\x00AIFF")
     r = upload_client.post(
         "/api/jobs",
-        files={"file": ("track.ogg", data, "audio/ogg")},
+        files={"file": ("track.aiff", data, "audio/aiff")},
     )
     assert r.status_code == 422
     assert "Unsupported file type" in r.json()["detail"]
@@ -183,6 +183,26 @@ def test_upload_flac_returns_job_id(upload_client):
     assert "job_id" in r.json()
 
 
+def test_upload_ogg_returns_job_id(upload_client):
+    data = io.BytesIO(b"OggS" + b"\x00" * 128)
+    r = upload_client.post(
+        "/api/jobs",
+        files={"file": ("my_track.ogg", data, "audio/ogg")},
+    )
+    assert r.status_code == 200
+    assert "job_id" in r.json()
+
+
+def test_upload_opus_returns_job_id(upload_client):
+    data = io.BytesIO(b"OggS" + b"\x00" * 128)
+    r = upload_client.post(
+        "/api/jobs",
+        files={"file": ("my_track.opus", data, "audio/opus")},
+    )
+    assert r.status_code == 200
+    assert "job_id" in r.json()
+
+
 # ─── Sections endpoint ────────────────────────────────────────────────────────
 
 

tests/test_stems_api.py

@@ -177,7 +177,7 @@ def test_all_stems_zip_rejects_bad_format(client, tmp_path):
     job.status = "done"
     _jobs[job.id] = job
     _make_stem_file(tmp_path, job.id, "vocals")
-    r = client.get(f"/api/jobs/{job.id}/stems/all.zip?format=ogg")
+    r = client.get(f"/api/jobs/{job.id}/stems/all.zip?format=aiff")
     assert r.status_code == 422
 
 
@@ -236,6 +236,41 @@ def test_all_stems_zip_mp3(client, tmp_path):
     assert len(zf.read("vocals.mp3")) > 0
 
 
+def test_all_stems_zip_ogg(client, tmp_path):
+    """OGG zip transcodes via ffmpeg (libvorbis); skip if ffmpeg isn't available."""
+    import io
+    import shutil
+    import zipfile
+
+    if shutil.which("ffmpeg") is None:
+        import pytest
+
+        pytest.skip("ffmpeg not available")
+
+    import struct
+
+    sr = 8000
+    nframes = sr // 10
+    data = b"\x00\x00" * nframes
+    hdr = b"RIFF" + struct.pack("<I", 36 + len(data)) + b"WAVE"
+    hdr += b"fmt " + struct.pack("<IHHIIHH", 16, 1, 1, sr, sr * 2, 2, 16)
+    hdr += b"data" + struct.pack("<I", len(data))
+    wav = hdr + data
+
+    job = Job(id="abcdefabcdb0")
+    job.status = "done"
+    job.title = "Track"
+    _jobs[job.id] = job
+    _make_stem_file(tmp_path, job.id, "vocals", wav)
+
+    r = client.get(f"/api/jobs/{job.id}/stems/all.zip?format=ogg")
+    assert r.status_code == 200
+    zf = zipfile.ZipFile(io.BytesIO(r.content))
+    assert zf.namelist() == ["vocals.ogg"]
+    ogg_bytes = zf.read("vocals.ogg")
+    assert ogg_bytes.startswith(b"OggS")
+
+
 # --- dynamic mixdown endpoint (#183) ---
 
 
@@ -262,7 +297,7 @@ def _done_job_with_stems(tmp_path, job_id: str, names) -> Job:
 
 
 def test_mixdown_rejects_bad_ext(client):
-    r = client.get("/api/jobs/abcdef000001/mixdown.ogg?stems=vocals&gains=1")
+    r = client.get("/api/jobs/abcdef000001/mixdown.aiff?stems=vocals&gains=1")
     assert r.status_code == 404