test(offline): make concurrency test deterministic and bounded

Replace the `sleep(0.15)` ordering hack with an explicit `threading.Event`
the fast thread sets in `finally`. The slow thread waits on that event
(bounded), then observes the flag — so we deterministically verify the
slow thread still sees offline mode after the fast thread has exited.

Also add timeouts to `barrier.wait()` and assert `not thread.is_alive()`
after the joins so the test can't hang on an unexpected failure path.

Co-Authored-By: Claude Opus 4.7 (1M context) <[email protected]>

.github/workflows/release.yml

@@ -26,45 +26,12 @@ jobs:
             args: ""
             python-version: "3.12"
             backend: "pytorch"
-          - platform: "ubuntu-22.04"
-            # --config override disables updater-artifact generation on Linux.
-            # tauri.conf.json has createUpdaterArtifacts: "v1Compatible" which
-            # on Linux wants to synthesize a .AppImage.tar.gz by downloading
-            # linuxdeploy at build time — this is what silently hangs CI
-            # (see v0.4.2 round 2, 25 min of no output after rpm bundling).
-            # We ship deb+rpm only; Linux users update via apt/dnf, not the
-            # Tauri in-app updater.
-            args: '--target x86_64-unknown-linux-gnu --bundles deb,rpm --verbose --config {"bundle":{"createUpdaterArtifacts":false}}'
-            python-version: "3.12"
-            backend: "pytorch"
 
     runs-on: ${{ matrix.platform }}
 
     steps:
       - uses: actions/checkout@v4
 
-      # Ubuntu runners ship with ~14 GB free; pip + PyInstaller + torch can
-      # peak well above that during the build. Reclaim ~25 GB by pruning
-      # preinstalled toolchains we don't use. This is what likely tripped
-      # the March 2026 Linux release attempts (see commit 103e98b
-      # "github runners suck") — not a code issue, a disk-pressure one.
-      - name: Free up disk space (ubuntu)
-        if: contains(matrix.platform, 'ubuntu') || contains(matrix.platform, 'namespace')
-        # Pinned to v1.3.1 (SHA) — this job runs with contents: write and
-        # handles signing secrets later, so we don't want a floating ref.
-        uses: jlumbroso/free-disk-space@54081f138730dfa15788a46383842cd2f914a1be
-        with:
-          tool-cache: false
-          android: true
-          dotnet: true
-          haskell: true
-          # large-packages: true would `apt-get remove '^llvm-.*'`, which
-          # cascade-removes reverse deps that won't be pulled back in by the
-          # `llvm-dev` install below. The other flags already free ~20 GB,
-          # enough for the Python + torch + PyInstaller build.
-          large-packages: false
-          swap-storage: true
-
       - name: Install dependencies (ubuntu only)
         if: contains(matrix.platform, 'ubuntu') || contains(matrix.platform, 'namespace')
         run: |
@@ -106,10 +73,8 @@ jobs:
           # fine on transformers 4.57.x in practice (verified in dev), so install
           # them --no-deps. mlx-audio's other runtime deps (huggingface_hub,
           # librosa, numpy, numba, pyloudnorm) are already in requirements.txt;
-          # miniaudio is in requirements-mlx.txt (needed by mlx_audio.stt,
-          # not transitively pulled by anything else — see issue #505); the
-          # rest (sounddevice, protobuf, sentencepiece, pyyaml, jinja2) are
-          # pulled in by other engines.
+          # the rest (sounddevice, miniaudio, protobuf, sentencepiece, pyyaml,
+          # jinja2) are pulled in by other engines.
           pip install --no-deps mlx-lm==0.31.1
           pip install --no-deps mlx-audio==0.4.1
 
@@ -168,21 +133,6 @@ jobs:
           p12-file-base64: ${{ secrets.APPLE_CERTIFICATE }}
           p12-password: ${{ secrets.APPLE_CERTIFICATE_PASSWORD }}
 
-      - name: Disk / environment snapshot (pre-bundle debug)
-        if: contains(matrix.platform, 'ubuntu') || contains(matrix.platform, 'namespace')
-        run: |
-          echo "=== df -h ==="
-          df -h
-          echo "=== free -h ==="
-          free -h
-          echo "=== Rust / Cargo ==="
-          rustc --version
-          cargo --version
-          echo "=== Bun ==="
-          bun --version
-          echo "=== Tauri CLI ==="
-          cd tauri && bun run tauri --version
-
       - name: Extract release notes from CHANGELOG.md
         id: changelog
         shell: bash
@@ -206,13 +156,7 @@ jobs:
             echo "CHANGELOG_EOF"
           } >> "$GITHUB_OUTPUT"
 
-      # Linux hang watchdog: previous releases silently wedged inside tauri
-      # bundling (possibly linuxdeploy/AppImage download, possibly cargo link).
-      # Cap the step at 30 min so we get logs instead of waiting out the 6hr
-      # job timeout. Other platforms historically complete in ~25 min, so 45
-      # is comfortable.
       - uses: tauri-apps/[email protected]
-        timeout-minutes: ${{ (contains(matrix.platform, 'ubuntu') || contains(matrix.platform, 'namespace')) && 30 || 45 }}
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
           TAURI_SIGNING_PRIVATE_KEY: ${{ secrets.TAURI_SIGNING_PRIVATE_KEY }}
@@ -224,9 +168,6 @@ jobs:
           APPLE_PROVIDER_SHORT_NAME: ${{ secrets.APPLE_PROVIDER_SHORT_NAME }}
           APPLE_API_ISSUER: ${{ secrets.APPLE_API_ISSUER }}
           APPLE_API_KEY: ${{ secrets.APPLE_API_KEY }}
-          # Stream subprocess stdout/stderr so the hang is visible in logs.
-          CARGO_TERM_VERBOSE: "true"
-          RUST_BACKTRACE: "1"
         with:
           projectPath: tauri
           tagName: v__VERSION__

app/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@voicebox/app",
-  "version": "0.4.2",
+  "version": "0.4.1",
   "private": true,
   "type": "module",
   "scripts": {

backend/backends/__init__.py

@@ -177,7 +177,7 @@ def _get_qwen_model_configs() -> list[ModelConfig]:
     backend_type = get_backend_type()
     if backend_type == "mlx":
         repo_1_7b = "mlx-community/Qwen3-TTS-12Hz-1.7B-Base-bf16"
-        repo_0_6b = "mlx-community/Qwen3-TTS-12Hz-0.6B-Base-bf16"
+        repo_0_6b = "mlx-community/Qwen3-TTS-12Hz-1.7B-Base-bf16"  # 0.6B not available in MLX, falls back
     else:
         repo_1_7b = "Qwen/Qwen3-TTS-12Hz-1.7B-Base"
         repo_0_6b = "Qwen/Qwen3-TTS-12Hz-0.6B-Base"

backend/backends/mlx_backend.py

@@ -45,9 +45,11 @@ class MLXTTSBackend:
         Returns:
             HuggingFace Hub model ID for MLX
         """
+        # MLX model mapping
         mlx_model_map = {
             "1.7B": "mlx-community/Qwen3-TTS-12Hz-1.7B-Base-bf16",
-            "0.6B": "mlx-community/Qwen3-TTS-12Hz-0.6B-Base-bf16",
+            # 0.6B not yet converted to MLX format
+            "0.6B": "mlx-community/Qwen3-TTS-12Hz-1.7B-Base-bf16",  # Fallback to 1.7B
         }
 
         if model_size not in mlx_model_map:

backend/requirements-mlx.txt

@@ -3,12 +3,6 @@
 
 mlx>=0.30.0
 
-# miniaudio is a runtime dep of mlx-audio's STT path (mlx_audio.stt).
-# mlx-audio itself is installed --no-deps (see comment below), so we
-# must list miniaudio explicitly here or transcription fails on fresh
-# M1 installs with `ModuleNotFoundError: miniaudio` (issue #505).
-miniaudio>=1.59
-
 # NOTE: mlx-audio is intentionally not listed here. From 0.3.1 onward it
 # declares `transformers==5.0.0rc3` / `>=5.0.0`, which conflicts with the
 # `transformers<=4.57.6` cap in requirements.txt and breaks CI's clean
@@ -16,7 +10,6 @@ miniaudio>=1.59
 # mlx_audio.stt.load) works fine on transformers 4.57.x in practice.
 #
 # Install it via `pip install --no-deps mlx-audio==0.4.1` after this file
-# (see .github/workflows/release.yml). Most other mlx-audio runtime deps
-# (huggingface_hub, librosa, mlx-lm, numba, numpy, protobuf, pyloudnorm,
-# sounddevice, tqdm) are already in requirements.txt or pulled in by
-# other engines.
+# (see .github/workflows/release.yml). All other mlx-audio runtime deps
+# (huggingface_hub, librosa, miniaudio, mlx-lm, numba, numpy, protobuf,
+# pyloudnorm, sounddevice, tqdm) are already in requirements.txt.

backend/tests/test_audio_preprocess.py

@@ -1,112 +0,0 @@
-"""
-Unit tests for reference-audio preprocessing.
-
-Covers :func:`backend.utils.audio.preprocess_reference_audio` and
-:func:`backend.utils.audio.validate_and_load_reference_audio`.
-"""
-
-import sys
-from pathlib import Path
-
-import numpy as np
-import pytest
-import soundfile as sf
-
-sys.path.insert(0, str(Path(__file__).parent.parent))
-
-from utils.audio import (  # noqa: E402
-    preprocess_reference_audio,
-    validate_and_load_reference_audio,
-)
-
-
-SR = 24000
-
-
-def _tone(duration_s: float, amp: float = 0.3, freq: float = 220.0) -> np.ndarray:
-    n = int(duration_s * SR)
-    t = np.arange(n, dtype=np.float32) / SR
-    return (amp * np.sin(2 * np.pi * freq * t)).astype(np.float32)
-
-
-def test_peak_cap_scales_hot_input():
-    audio = _tone(3.0, amp=0.99)
-    out = preprocess_reference_audio(audio, SR)
-    assert np.abs(out).max() <= 0.951
-
-
-def test_peak_cap_leaves_moderate_input_untouched():
-    audio = _tone(3.0, amp=0.5)
-    out = preprocess_reference_audio(audio, SR)
-    assert np.isclose(np.abs(out).max(), 0.5, atol=1e-3)
-
-
-def test_dc_offset_removed():
-    audio = _tone(3.0, amp=0.3) + 0.1
-    out = preprocess_reference_audio(audio, SR)
-    assert abs(float(np.mean(out))) < 1e-3
-
-
-def test_silence_is_trimmed_with_padding_kept():
-    silence = np.zeros(int(SR * 1.0), dtype=np.float32)
-    speech = _tone(3.0, amp=0.3)
-    audio = np.concatenate([silence, speech, silence])
-    out = preprocess_reference_audio(audio, SR)
-    # Most of the 2s of leading/trailing silence should be gone, but the
-    # 3s of speech plus ~200ms of padding should remain.
-    assert len(audio) - len(out) >= SR, "expected >=1s of silence trimmed"
-    assert len(out) >= int(3.0 * SR), "speech body should be preserved"
-
-
-def test_clean_audio_is_not_padded_past_original_length():
-    # Well-recorded audio with no edge silence shouldn't get longer after
-    # preprocessing — otherwise a 29.9 s upload could be pushed past the
-    # 30 s max_duration ceiling downstream.
-    audio = _tone(3.0, amp=0.3)
-    out = preprocess_reference_audio(audio, SR)
-    assert len(out) <= len(audio)
-
-
-def test_empty_input_returns_empty():
-    out = preprocess_reference_audio(np.zeros(0, dtype=np.float32), SR)
-    assert out.size == 0
-
-
-def test_validate_accepts_previously_rejected_hot_file(tmp_path):
-    audio = _tone(3.0, amp=0.995)
-    path = tmp_path / "hot.wav"
-    sf.write(str(path), audio, SR)
-
-    ok, err, out_audio, out_sr = validate_and_load_reference_audio(str(path))
-
-    assert ok, f"expected pass, got error: {err}"
-    assert out_audio is not None
-    assert out_sr == SR
-    assert np.abs(out_audio).max() <= 0.951
-
-
-def test_validate_still_rejects_silent_input(tmp_path):
-    audio = np.zeros(int(SR * 3.0), dtype=np.float32)
-    path = tmp_path / "silent.wav"
-    sf.write(str(path), audio, SR)
-
-    ok, err, _, _ = validate_and_load_reference_audio(str(path))
-
-    assert not ok
-    assert err is not None
-    assert "too short" in err.lower() or "quiet" in err.lower()
-
-
-def test_validate_rejects_too_short(tmp_path):
-    audio = _tone(0.5, amp=0.3)
-    path = tmp_path / "short.wav"
-    sf.write(str(path), audio, SR)
-
-    ok, err, _, _ = validate_and_load_reference_audio(str(path))
-
-    assert not ok
-    assert "too short" in (err or "").lower()
-
-
-if __name__ == "__main__":
-    pytest.main([__file__, "-v"])

backend/tests/test_qwen_download.py

@@ -175,7 +175,7 @@ async def main():
     print("   ✅ Server is running")
 
     # Test model
-    model_name = "qwen-tts-0.6B"
+    model_name = "qwen-tts-0.6B"  # Note: 0.6B currently maps to 1.7B on MLX
     
     # Check current status
     print(f"\n📊 Checking status of {model_name}...")

backend/utils/audio.py

@@ -199,66 +199,6 @@ def trim_tts_output(
     return trimmed
 
 
-def preprocess_reference_audio(
-    audio: np.ndarray,
-    sample_rate: int,
-    peak_target: float = 0.95,
-    trim_top_db: float = 40.0,
-    edge_padding_ms: int = 100,
-) -> np.ndarray:
-    """
-    Clean up a reference-audio sample before validation/storage.
-
-    Removes DC offset, trims leading/trailing silence, and caps the peak so a
-    slightly-hot recording doesn't get rejected downstream as "clipping". The
-    goal is to accept reasonable real-world recordings — not to repair badly
-    distorted ones. True clipping artifacts inside the waveform can't be
-    recovered by peak scaling and will still sound bad.
-
-    Args:
-        audio: Mono audio array.
-        sample_rate: Sample rate of ``audio`` in Hz.
-        peak_target: Peak amplitude cap in [0, 1]. Applied only if the input
-            peak exceeds this value.
-        trim_top_db: Silence threshold for edge trimming, in dB below peak.
-            40 dB sits below normal speech dynamic range (≈30 dB) so soft
-            trailing syllables are preserved, while still catching obvious
-            leading/trailing silence. Lower values are more aggressive;
-            librosa's own default is 60.
-        edge_padding_ms: Milliseconds of padding to add back at each edge
-            *only if* trimming shortened the waveform, so TTS engines have a
-            brief silence to anchor on without ever making the output longer
-            than the input.
-
-    Returns:
-        Preprocessed audio array (float32).
-    """
-    audio = audio.astype(np.float32, copy=False)
-
-    if audio.size == 0:
-        return audio
-
-    audio = audio - float(np.mean(audio))
-
-    trimmed, _ = librosa.effects.trim(audio, top_db=trim_top_db)
-    if 0 < trimmed.size < audio.size:
-        pad_each = int(sample_rate * edge_padding_ms / 1000)
-        # Never pad past the original length — for near-max-duration uploads
-        # an unconditional pad would push them over the 30 s ceiling and
-        # trigger a spurious "too long" rejection.
-        headroom = (audio.size - trimmed.size) // 2
-        pad = min(pad_each, max(headroom, 0))
-        if pad > 0:
-            trimmed = np.pad(trimmed, (pad, pad), mode="constant")
-        audio = trimmed
-
-    peak = float(np.abs(audio).max())
-    if peak > peak_target and peak > 0:
-        audio = audio * (peak_target / peak)
-
-    return audio
-
-
 def validate_reference_audio(
     audio_path: str,
     min_duration: float = 2.0,
@@ -267,13 +207,13 @@ def validate_reference_audio(
 ) -> Tuple[bool, Optional[str]]:
     """
     Validate reference audio for voice cloning.
-
+    
     Args:
         audio_path: Path to audio file
         min_duration: Minimum duration in seconds
         max_duration: Maximum duration in seconds
         min_rms: Minimum RMS level
-
+        
     Returns:
         Tuple of (is_valid, error_message)
     """
@@ -291,28 +231,26 @@ def validate_and_load_reference_audio(
 ) -> Tuple[bool, Optional[str], Optional[np.ndarray], Optional[int]]:
     """
     Validate and load reference audio in a single pass.
-
-    Applies :func:`preprocess_reference_audio` before checks so that
-    slightly-hot recordings aren't rejected as clipping. Duration and RMS
-    checks run on the preprocessed waveform.
-
+    
     Returns:
         Tuple of (is_valid, error_message, audio_array, sample_rate)
     """
     try:
         audio, sr = load_audio(audio_path)
-        audio = preprocess_reference_audio(audio, sr)
         duration = len(audio) / sr
-
+        
         if duration < min_duration:
             return False, f"Audio too short (minimum {min_duration} seconds)", None, None
         if duration > max_duration:
             return False, f"Audio too long (maximum {max_duration} seconds)", None, None
-
+        
         rms = np.sqrt(np.mean(audio**2))
         if rms < min_rms:
             return False, "Audio is too quiet or silent", None, None
-
+        
+        if np.abs(audio).max() > 0.99:
+            return False, "Audio is clipping (reduce input gain)", None, None
+        
         return True, None, audio, sr
     except Exception as e:
         return False, f"Error validating audio: {str(e)}", None, None

docs/PROJECT_STATUS.md

@@ -309,7 +309,11 @@ Still reported. Users get stuck downloads, can't resume, offline mode edge cases
 
 **Fix path:** PR #443 addresses infinite offline retry. CustomVoice-specific download failures (#475, #445) need triage — likely related to frozen-binary import fixes in PR #438. TADA cluster (#336, #348) and macOS ARM import regressions (#287, #275, #304) need a dedicated triage pass.
 
-**Qwen 0.6B-downloads-1.7B reports:** **#485** (2026-04-19), **#423** (macOS M1), **#329**. Originally a stale-fallback bug: `mlx-community/Qwen3-TTS-12Hz-0.6B-Base-bf16` wasn't published when MLX support shipped, so the 0.6B slot was aliased to the 1.7B repo. The 0.6B bf16 conversion is live now and both `backend/backends/mlx_backend.py` and `backend/backends/__init__.py` point at their correct repos. Qwen CustomVoice is unaffected — it runs via PyTorch on all platforms, both sizes always have dedicated repos.
+**Qwen 0.6B-downloads-1.7B reports:** **#485** (2026-04-19), **#423** (macOS M1), **#329**. Platform-dependent:
+
+- **On MLX (Apple Silicon) — not a bug.** `mlx-community` only publishes 1.7B-Base-bf16 weights, so the 0.6B Base option intentionally resolves to the same repo (`backend/backends/__init__.py:180` — `# 0.6B not available in MLX, falls back`). UX gap: the selector offers a size that doesn't exist on the active backend. Fix: (a) hide the 0.6B option on MLX, or (b) label it "0.6B (uses 1.7B on Apple Silicon)".
+- **On PyTorch (Windows/Linux/CUDA/ROCm/XPU/CPU) — real bug if reported.** Both 0.6B and 1.7B have distinct repos (`Qwen/Qwen3-TTS-12Hz-0.6B-Base` vs `-1.7B-Base`). Triage each report by platform before merging into the MLX cluster.
+- **Qwen CustomVoice (either platform)** — no fallback, both sizes always have dedicated repos.
 
 ### Language Requests (ongoing)
 
@@ -389,7 +393,7 @@ Notable:
 | **#306** ("voice model"), **#389** ("New model"), **#473** ("New functionality") | Title-only issues, no content. Request details or close. |
 | **#309** | Uninstall/cleanup question. Answer and close. |
 | **#241** | "How to use in Colab" — support question, not a bug. |
-| **#423** / **#485** / **#329** | Stale MLX fallback to 1.7B repo — fixed; 0.6B bf16 conversion now live on `mlx-community`, registry points at correct repo on both backends. |
+| **#423** / **#485** / **#329** | Platform-dependent. On MLX: not a bug (0.6B weights don't exist upstream, fallback is intentional — fix UX). On PyTorch: real bug if reproducible. Classify each by reporter's platform before deduping. |
 | **#336** / **#348** | TADA download/registration cluster — triage together. |
 | **#287** / **#275** / **#304** | macOS ARM import regressions on new version — likely one root cause. |
 | **#292**, **#349** | Possibly already fixed by merged PRs (#321/#412 and #345). Verify + close. |

docs/content/docs/overview/troubleshooting.mdx

@@ -446,6 +446,20 @@ Restart the app to create a fresh database.
   huggingface-cli download Qwen/Qwen3-TTS-12Hz-1.7B-Base
   ```
 
+### Qwen 0.6B Downloads the Same Files as 1.7B on Apple Silicon
+
+**Symptoms:**
+- You select Qwen 0.6B on an Apple Silicon Mac and the download is the same size as 1.7B
+- Generation speed and VRAM usage match 1.7B, not the expected smaller model
+
+**Explanation:**
+This is intentional, not a bug. The MLX community only publishes `mlx-community/Qwen3-TTS-12Hz-1.7B-Base-bf16` — there is no 0.6B MLX build. Voicebox's model registry falls back to the 1.7B weights when 0.6B is selected on MLX (see `backend/backends/__init__.py`).
+
+**Solution:**
+- On Apple Silicon, both size options use the 1.7B model — pick either.
+- If you specifically need a smaller model, switch to **Kokoro 82M** (~350 MB) or **LuxTTS** (~300 MB) — both CPU-realtime.
+- On Windows/Linux with PyTorch, 0.6B and 1.7B are distinct repos and behave differently.
+
 ### Wrong Model Version
 
 **Symptoms:**

landing/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@voicebox/landing",
-  "version": "0.4.2",
+  "version": "0.4.1",
   "description": "Landing page for voicebox.sh",
   "scripts": {
     "dev": "next dev --turbo",

package.json

@@ -1,6 +1,6 @@
 {
   "name": "voicebox",
-  "version": "0.4.2",
+  "version": "0.4.1",
   "private": true,
   "workspaces": [
     "app",

tauri/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@voicebox/tauri",
   "private": true,
-  "version": "0.4.2",
+  "version": "0.4.1",
   "type": "module",
   "scripts": {
     "dev": "vite",

tauri/src-tauri/Cargo.lock

@@ -5041,7 +5041,7 @@ checksum = "0b928f33d975fc6ad9f86c8f283853ad26bdd5b10b7f1542aa2fa15e2289105a"
 
 [[package]]
 name = "voicebox"
-version = "0.4.2"
+version = "0.4.1"
 dependencies = [
  "base64 0.22.1",
  "core-foundation-sys",

tauri/src-tauri/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "voicebox"
-version = "0.4.2"
+version = "0.4.1"
 description = "A production-quality desktop app for Qwen3-TTS voice cloning and generation"
 authors = ["you"]
 license = ""

tauri/src-tauri/tauri.conf.json

@@ -1,7 +1,7 @@
 {
   "$schema": "https://schema.tauri.app/config/2",
   "productName": "Voicebox",
-  "version": "0.4.2",
+  "version": "0.4.1",
   "identifier": "sh.voicebox.app",
   "build": {
     "beforeDevCommand": "bun run dev",

web/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@voicebox/web",
   "private": true,
-  "version": "0.4.2",
+  "version": "0.4.1",
   "type": "module",
   "scripts": {
     "dev": "vite",