fix(worker): preserve isolated OmniVoice synthesis settings

backend/worker/executor.py

@@ -208,9 +208,11 @@ class TaskExecutor:
         from services.text_normalization import normalize_for_tts
 
         text = normalize_for_tts(row.get("text") or "", row.get("language"))
+        seed = None
         if row.get("seed") is not None:
             import torch
-            torch.manual_seed(int(row["seed"]))
+            seed = int(row["seed"])
+            torch.manual_seed(seed)
         kwargs = {
             "language": row.get("language") if row.get("language") != "Auto" else None,
             "ref_audio": row.get("ref_audio"), "ref_text": row.get("ref_text"),
@@ -221,6 +223,11 @@ class TaskExecutor:
             "speed": float(row.get("speed") or 1.0), "denoise": True,
             "postprocess_output": True,
         }
+        if (
+            getattr(backend, "supports_native_omnivoice_controls", False)
+            and seed is not None
+        ):
+            kwargs["seed"] = seed
         audio = backend.generate(text=text, **kwargs)
         preset = row.get("effect_preset") or "broadcast"
         if preset != "raw":
@@ -332,7 +339,10 @@ class TaskExecutor:
             key: value for key, value in opts.to_manifest().items()
             if value is not None and key not in ("seed", "vary_repeats")
         }
-        if isinstance(backend, OmniVoiceBackend):
+        native_proxy = bool(
+            getattr(backend, "supports_native_omnivoice_controls", False)
+        )
+        if isinstance(backend, OmniVoiceBackend) or native_proxy:
             extra.setdefault("num_step", 32)
             extra.setdefault("guidance_scale", 2.0)
             for key in ("emo_vector", "emo_text", "emo_alpha"):
@@ -341,11 +351,13 @@ class TaskExecutor:
         def synth(text, index, speed=None):
             voice = voices[int(index)]
             base_seed = opts.seed if opts.seed is not None else voice.get("seed")
+            seed = None
             if base_seed is not None:
                 import torch
                 nonce = occurrence["value"] if opts.vary_repeats else 0
                 occurrence["value"] += 1
-                torch.manual_seed(segment_seed(base_seed, text, nonce))
+                seed = segment_seed(base_seed, text, nonce)
+                torch.manual_seed(seed)
             kwargs = {
                 "language": language,
                 "ref_audio": voice.get("ref_audio"),
@@ -354,6 +366,8 @@ class TaskExecutor:
                 "speed": float(speed) if speed else 1.0,
                 **extra,
             }
+            if native_proxy and seed is not None:
+                kwargs["seed"] = seed
             return backend.generate(text, **kwargs)
 
         spans = [Span(voice_id=str(i), text=row.get("text", ""),

docs/engines/omnivoice-subprocess.md

@@ -32,7 +32,7 @@ lower call overhead.
 
 ## Selecting it
 
-- **Settings -> Engines**, or
+- **Model Catalogue → Engines**, or
 - `OMNIVOICE_TTS_BACKEND=omnivoice-subprocess`
 
 The explicit engine is opt-in on CUDA, ROCm, and CPU. Apple Silicon gets the

tests/test_audiobook_remote.py

@@ -54,3 +54,30 @@ def test_audiobook_worker_marks_and_encodes_chapter(monkeypatch):
     )
     assert len(audio) == 20
     assert marked == [(100, "worker.executor.tts")]
+
+
+def test_audiobook_worker_forwards_mps_proxy_quality_and_seed():
+    import numpy as np
+    from services.audiobook import segment_seed
+    from worker.executor import TaskExecutor
+
+    calls = []
+
+    class Backend:
+        sample_rate = 100
+        supports_native_omnivoice_controls = True
+
+        def generate(self, text, **kwargs):
+            calls.append((text, kwargs))
+            return np.ones(20, dtype=np.float32)
+
+    TaskExecutor._synthesize_audiobook(
+        Backend(), [{"text": "hello", "pause_ms_after": 0}],
+        [{"ref_text": None, "instruct": None, "seed": 42}],
+        {"ref_audio": [None], "expressive": {}, "watermark": False},
+    )
+
+    _text, kwargs = calls[0]
+    assert kwargs["num_step"] == 32
+    assert kwargs["guidance_scale"] == 2.0
+    assert kwargs["seed"] == segment_seed(42, "hello")

tests/test_worker_executor_residency.py

@@ -83,6 +83,27 @@ class _FakeBackend:
         type(self).unloaded += 1
 
 
+def test_dub_worker_forwards_seed_into_mps_proxy():
+    calls = []
+
+    class Proxy:
+        sample_rate = 24_000
+        applies_own_mastering = False
+        supports_native_omnivoice_controls = True
+
+        def generate(self, text, **kwargs):
+            calls.append((text, kwargs))
+            import torch
+
+            return torch.zeros(1, 240)
+
+    TaskExecutor._synthesize_dub_segment(Proxy(), {
+        "text": "hello", "seed": 123, "effect_preset": "raw",
+    })
+
+    assert calls[0][1]["seed"] == 123
+
+
 @pytest.mark.asyncio
 async def test_cancelling_execution_drains_the_blocking_engine_thread(monkeypatch):
     started = threading.Event()