Compare commits

...
18 Commits
Author SHA1 Message Date
James Pine 34e17bd469 Fix LuxTTS + Chatterbox in prod: bundle espeak/perth data, fix multiprocessing
- collect-all piper_phonemize to bundle espeak-ng-data for LuxTTS phonemization
- Set ESPEAK_DATA_PATH in frozen builds so the C library finds bundled data
- collect-all perth to bundle pretrained watermark model for Chatterbox
- Add multiprocessing.freeze_support() to fix resource_tracker subprocess crash
2026-03-15 16:02:09 -07:00
James Pine aada13a5c9 Collect all inflect files for PyInstaller (fixes typeguard inspect.getsource) 2026-03-15 14:32:10 -07:00
James Pine de8558d197 Fix prod build: download progress, robust stderr, full tracebacks
- Force tqdm disable=False in TrackedTqdm so byte progress works in prod
  (huggingface_hub disables tqdm based on logger level, which prevents
  self.n from updating — our progress tracking needs the counter even
  though we don't render to terminal)
- Harden devnull redirect to test writability, not just None check
- Add full traceback logging to all backend error handlers
- Add chatterbox/luxtts/zipvoice hidden imports and metadata to spec
2026-03-15 14:23:11 -07:00
Jamie Pine 9d79ea367a Only use --noconsole on Windows, macOS/Linux need stdout for Tauri logs 2026-03-15 12:07:35 -07:00
Jamie Pine 04316f7adc Copy metadata for requests/transformers/huggingface-hub to fix PyInstaller metadata lookup 2026-03-15 11:35:05 -07:00
Jamie Pine 4e4361d350 Fix noconsole crash: redirect None stdout/stderr to devnull on Windows 2026-03-15 11:27:35 -07:00
Jamie Pine e9a249587c Collect all linacodec files for PyInstaller (fixes inspect.getsource in Vocos) 2026-03-15 11:06:13 -07:00
Jamie Pine d8a9ed7d15 Enable updater artifacts with v1Compatible for tauri-action sig generation 2026-03-15 10:54:12 -07:00
Jamie Pine 3dbf1c200e Revert "Bump version: 0.2.3 → 0.2.4"
This reverts commit 40fcb8d917.
2026-03-15 10:20:31 -07:00
Jamie Pine 40fcb8d917 Bump version: 0.2.3 → 0.2.4 2026-03-15 10:18:51 -07:00
Jamie Pine ad64d1c3d9 Collect all zipvoice files for PyInstaller (fixes source code error) 2026-03-15 10:18:40 -07:00
Jamie Pine f826e45250 Install chatterbox-tts in CI release workflow 2026-03-15 10:17:23 -07:00
Jamie Pine 3d53c06c5b Bump version: 0.2.2 → 0.2.3 2026-03-15 10:08:56 -07:00
James Pine 9835b9f6d4 fix: prevent stale release data by removing Next.js fetch cache
Replace next: { revalidate: 600 } with cache: 'no-store' on GitHub
API fetches so new releases show up within 5 minutes (in-memory cache
only, no Next.js/Vercel cache layer on top).
2026-03-15 10:07:50 -07:00
Jamie Pine a15dd30b1e Update tauri-action to v0.6 to fix updater JSON and signature generation 2026-03-15 10:05:36 -07:00
Jamie Pine 1d343ac071 Treat missing/draft releases as up-to-date instead of showing error 2026-03-15 09:52:17 -07:00
James Pine ca602de0ae fix: don't reset audio player when unmuting during playback 2026-03-15 09:29:44 -07:00
James Pine cdc0293ca8 feat: add /linux-install page with build-from-source instructions
Linux download card now links to /linux-install instead of a direct
binary download. The page explains the CI situation and gives
clone + setup + build commands.
2026-03-15 09:17:30 -07:00
23 changed files with 274 additions and 24 deletions
+1 -1
View File
@@ -1,5 +1,5 @@
[bumpversion]
current_version = 0.2.2
current_version = 0.2.3
commit = True
tag = True
tag_name = v{new_version}
+3 -1
View File
@@ -61,6 +61,7 @@ jobs:
python -m pip install --upgrade pip
pip install pyinstaller
pip install -r backend/requirements.txt
pip install --no-deps chatterbox-tts
- name: Install MLX dependencies (Apple Silicon only)
if: matrix.backend == 'mlx'
@@ -122,7 +123,7 @@ jobs:
p12-file-base64: ${{ secrets.APPLE_CERTIFICATE }}
p12-password: ${{ secrets.APPLE_CERTIFICATE_PASSWORD }}
- uses: tauri-apps/tauri-action@v0
- uses: tauri-apps/tauri-action@v0.6
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
TAURI_SIGNING_PRIVATE_KEY: ${{ secrets.TAURI_SIGNING_PRIVATE_KEY }}
@@ -173,6 +174,7 @@ jobs:
python -m pip install --upgrade pip
pip install pyinstaller
pip install -r backend/requirements.txt
pip install --no-deps chatterbox-tts
- name: Install PyTorch with CUDA 12.1
run: |
+1 -1
View File
@@ -1,6 +1,6 @@
{
"name": "@voicebox/app",
"version": "0.2.2",
"version": "0.2.3",
"private": true,
"type": "module",
"scripts": {
+1 -1
View File
@@ -1,3 +1,3 @@
# Backend package
__version__ = "0.2.2"
__version__ = "0.2.3"
+2 -1
View File
@@ -224,7 +224,8 @@ class ChatterboxTTSBackend:
task_manager.error_download(model_name, str(e))
raise
except Exception as e:
logger.error(f"Failed to load Chatterbox: {e}")
import traceback
logger.error(f"Failed to load Chatterbox: {e}\n{traceback.format_exc()}")
if not is_cached:
progress_manager.mark_error(model_name, str(e))
task_manager.error_download(model_name, str(e))
+2 -1
View File
@@ -228,7 +228,8 @@ class ChatterboxTurboTTSBackend:
task_manager.error_download(model_name, str(e))
raise
except Exception as e:
logger.error(f"Failed to load Chatterbox Turbo: {e}")
import traceback
logger.error(f"Failed to load Chatterbox Turbo: {e}\n{traceback.format_exc()}")
if not is_cached:
progress_manager.mark_error(model_name, str(e))
task_manager.error_download(model_name, str(e))
+2 -1
View File
@@ -149,7 +149,8 @@ class LuxTTSBackend:
logger.info("LuxTTS loaded successfully")
except Exception as e:
logger.error(f"Failed to load LuxTTS: {e}")
import traceback
logger.error(f"Failed to load LuxTTS: {e}\n{traceback.format_exc()}")
if not is_cached:
progress_manager.mark_error(model_name, str(e))
task_manager.error_download(model_name, str(e))
+23 -1
View File
@@ -34,10 +34,14 @@ def build_server(cuda=False):
args = [
'server.py', # Use server.py as entry point instead of main.py
'--onefile',
'--noconsole', # No visible console window on Windows
'--name', binary_name,
]
# Hide console window on Windows only. On macOS/Linux the sidecar needs
# stdout/stderr for Tauri to capture logs.
if platform.system() == "Windows":
args.append('--noconsole')
# Add local qwen_tts path if specified (for editable installs)
qwen_tts_path = os.getenv('QWEN_TTS_PATH')
if qwen_tts_path and Path(qwen_tts_path).exists():
@@ -76,6 +80,8 @@ def build_server(cuda=False):
'--hidden-import', 'backend.backends.luxtts_backend',
'--hidden-import', 'zipvoice',
'--hidden-import', 'zipvoice.luxvoice',
'--collect-all', 'zipvoice',
'--collect-all', 'linacodec',
'--hidden-import', 'torch',
'--hidden-import', 'transformers',
'--hidden-import', 'fastapi',
@@ -90,11 +96,27 @@ def build_server(cuda=False):
'--hidden-import', 'qwen_tts.core',
'--hidden-import', 'qwen_tts.cli',
'--copy-metadata', 'qwen-tts',
'--copy-metadata', 'requests',
'--copy-metadata', 'transformers',
'--copy-metadata', 'huggingface-hub',
'--copy-metadata', 'tokenizers',
'--copy-metadata', 'safetensors',
'--copy-metadata', 'tqdm',
'--hidden-import', 'requests',
'--collect-submodules', 'qwen_tts',
'--collect-data', 'qwen_tts',
# Fix for pkg_resources and jaraco namespace packages
'--hidden-import', 'pkg_resources.extern',
'--collect-submodules', 'jaraco',
# inflect uses typeguard @typechecked which calls inspect.getsource()
# at import time — needs .py source files, not just .pyc bytecode
'--collect-all', 'inflect',
# perth ships pretrained watermark model files (hparams.yaml, .pth.tar)
# in perth/perth_net/pretrained/ — needed by chatterbox at runtime
'--collect-all', 'perth',
# piper_phonemize ships espeak-ng-data/ (phoneme tables, language dicts)
# needed by LuxTTS for text-to-phoneme conversion
'--collect-all', 'piper_phonemize',
])
# Add CUDA-specific hidden imports
+33
View File
@@ -6,6 +6,39 @@ absolute imports instead of relative imports.
"""
import sys
import os
# On Windows with --noconsole (PyInstaller), sys.stdout/stderr are None.
# They can also be broken file objects in some edge cases.
# Redirect to devnull to prevent crashes from print()/tqdm/logging.
def _is_writable(stream):
"""Check if a stream is usable for writing."""
if stream is None:
return False
try:
stream.write("")
return True
except Exception:
return False
if not _is_writable(sys.stdout):
sys.stdout = open(os.devnull, 'w')
if not _is_writable(sys.stderr):
sys.stderr = open(os.devnull, 'w')
# PyInstaller + multiprocessing: child processes re-execute the frozen binary
# with internal arguments. freeze_support() handles this and exits early.
import multiprocessing
multiprocessing.freeze_support()
# In frozen builds, piper_phonemize's espeak-ng C library falls back to
# /usr/share/espeak-ng-data/ which doesn't exist. Point it at the bundled
# data directory instead.
if getattr(sys, 'frozen', False):
_meipass = getattr(sys, '_MEIPASS', os.path.dirname(sys.executable))
_espeak_data = os.path.join(_meipass, 'piper_phonemize', 'espeak-ng-data')
if os.path.isdir(_espeak_data):
os.environ.setdefault('ESPEAK_DATA_PATH', _espeak_data)
# Fast path: handle --version before any heavy imports so the Rust
# version check doesn't block for 30+ seconds loading torch etc.
+6
View File
@@ -64,11 +64,17 @@ class HFProgressTracker:
if key in tqdm_kwargs:
filtered_kwargs[key] = value
# Force-enable the progress bar — we're tracking progress ourselves,
# we don't need tqdm to render to a terminal, but we DO need
# self.n to be updated when update() is called.
filtered_kwargs['disable'] = False
# Try to initialize with filtered kwargs, fall back to all kwargs if that fails
try:
super().__init__(*args, **filtered_kwargs)
except TypeError:
# If filtering failed, try with all kwargs (maybe tqdm version accepts them)
kwargs['disable'] = False
super().__init__(*args, **kwargs)
self._tracker_filename = filename or "unknown"
+11 -1
View File
@@ -6,13 +6,23 @@ from PyInstaller.utils.hooks import copy_metadata
datas = []
binaries = []
hiddenimports = ['backend', 'backend.main', 'backend.config', 'backend.database', 'backend.models', 'backend.profiles', 'backend.history', 'backend.tts', 'backend.transcribe', 'backend.platform_detect', 'backend.backends', 'backend.backends.pytorch_backend', 'backend.utils.audio', 'backend.utils.cache', 'backend.utils.progress', 'backend.utils.hf_progress', 'backend.utils.validation', 'backend.cuda_download', 'backend.effects', 'backend.utils.effects', 'backend.versions', 'pedalboard', 'torch', 'transformers', 'fastapi', 'uvicorn', 'sqlalchemy', 'librosa', 'soundfile', 'qwen_tts', 'qwen_tts.inference', 'qwen_tts.inference.qwen3_tts_model', 'qwen_tts.inference.qwen3_tts_tokenizer', 'qwen_tts.core', 'qwen_tts.cli', 'pkg_resources.extern', 'backend.backends.mlx_backend', 'mlx', 'mlx.core', 'mlx.nn', 'mlx_audio', 'mlx_audio.tts', 'mlx_audio.stt']
hiddenimports = ['backend', 'backend.main', 'backend.config', 'backend.database', 'backend.models', 'backend.profiles', 'backend.history', 'backend.tts', 'backend.transcribe', 'backend.platform_detect', 'backend.backends', 'backend.backends.pytorch_backend', 'backend.utils.audio', 'backend.utils.cache', 'backend.utils.progress', 'backend.utils.hf_progress', 'backend.utils.validation', 'backend.cuda_download', 'backend.effects', 'backend.utils.effects', 'backend.versions', 'pedalboard', 'chatterbox', 'chatterbox.tts_turbo', 'chatterbox.mtl_tts', 'backend.backends.chatterbox_backend', 'backend.backends.chatterbox_turbo_backend', 'backend.backends.luxtts_backend', 'zipvoice', 'zipvoice.luxvoice', 'torch', 'transformers', 'fastapi', 'uvicorn', 'sqlalchemy', 'librosa', 'soundfile', 'qwen_tts', 'qwen_tts.inference', 'qwen_tts.inference.qwen3_tts_model', 'qwen_tts.inference.qwen3_tts_tokenizer', 'qwen_tts.core', 'qwen_tts.cli', 'requests', 'pkg_resources.extern', 'backend.backends.mlx_backend', 'mlx', 'mlx.core', 'mlx.nn', 'mlx_audio', 'mlx_audio.tts', 'mlx_audio.stt']
datas += collect_data_files('qwen_tts')
datas += copy_metadata('qwen-tts')
datas += copy_metadata('requests')
datas += copy_metadata('transformers')
datas += copy_metadata('huggingface-hub')
datas += copy_metadata('tokenizers')
datas += copy_metadata('safetensors')
datas += copy_metadata('tqdm')
hiddenimports += collect_submodules('qwen_tts')
hiddenimports += collect_submodules('jaraco')
hiddenimports += collect_submodules('mlx')
hiddenimports += collect_submodules('mlx_audio')
tmp_ret = collect_all('zipvoice')
datas += tmp_ret[0]; binaries += tmp_ret[1]; hiddenimports += tmp_ret[2]
tmp_ret = collect_all('linacodec')
datas += tmp_ret[0]; binaries += tmp_ret[1]; hiddenimports += tmp_ret[2]
tmp_ret = collect_all('mlx')
datas += tmp_ret[0]; binaries += tmp_ret[1]; hiddenimports += tmp_ret[2]
tmp_ret = collect_all('mlx_audio')
+1 -1
View File
@@ -1,6 +1,6 @@
{
"name": "@voicebox/landing",
"version": "0.2.2",
"version": "0.2.3",
"description": "Landing page for voicebox.sh",
"scripts": {
"dev": "bun --bun next dev --turbo",
-1
View File
@@ -2,7 +2,6 @@ import { NextResponse } from 'next/server';
import { getLatestRelease } from '@/lib/releases';
export const dynamic = 'force-dynamic';
export const revalidate = 600; // Revalidate every 10 minutes
export async function GET() {
try {
+169
View File
@@ -0,0 +1,169 @@
import type { Metadata } from 'next';
import { Footer } from '@/components/Footer';
import { Navbar } from '@/components/Navbar';
import { GITHUB_REPO } from '@/lib/constants';
export const metadata: Metadata = {
title: 'Linux Install - Voicebox',
description: 'Build Voicebox from source on Linux. Clone, setup, and build in three commands.',
};
export default function LinuxInstall() {
return (
<>
<Navbar />
<section className="relative pt-32 pb-24">
<div className="mx-auto max-w-2xl px-6">
<h1 className="text-3xl font-bold tracking-tight text-foreground">Install on Linux</h1>
<p className="mt-4 text-muted-foreground">
We&apos;re currently working through CI issues that prevent us from shipping a reliable
pre-built binary for Linux. In the meantime, building from source is straightforward and
takes just a few minutes.
</p>
<div className="mt-10 space-y-6">
{/* Prerequisites */}
<div>
<h2 className="text-sm font-medium text-muted-foreground uppercase tracking-wider mb-3">
Prerequisites
</h2>
<ul className="list-disc list-inside text-sm text-muted-foreground space-y-1">
<li>
<a
href="https://git-scm.com"
target="_blank"
rel="noopener noreferrer"
className="text-foreground hover:underline"
>
Git
</a>
</li>
<li>
<a
href="https://www.rust-lang.org/tools/install"
target="_blank"
rel="noopener noreferrer"
className="text-foreground hover:underline"
>
Rust
</a>
</li>
<li>
<a
href="https://github.com/casey/just#installation"
target="_blank"
rel="noopener noreferrer"
className="text-foreground hover:underline"
>
just
</a>{' '}
— install via{' '}
<code className="text-xs bg-muted px-1.5 py-0.5 rounded">cargo install just</code>
</li>
<li>
<a
href="https://bun.sh"
target="_blank"
rel="noopener noreferrer"
className="text-foreground hover:underline"
>
Bun
</a>
</li>
<li>
Tauri system deps —{' '}
<a
href="https://v2.tauri.app/start/prerequisites/#linux"
target="_blank"
rel="noopener noreferrer"
className="text-foreground hover:underline"
>
see Tauri docs
</a>
</li>
</ul>
</div>
{/* Steps */}
<div>
<h2 className="text-sm font-medium text-muted-foreground uppercase tracking-wider mb-3">
Build from source
</h2>
<div className="space-y-3">
<div className="rounded-lg border border-border bg-card/60 p-4 font-mono text-sm">
<div className="text-muted-foreground select-none"># Clone the repo</div>
<div>git clone https://github.com/jamiepine/voicebox.git</div>
<div>cd voicebox</div>
</div>
<div className="rounded-lg border border-border bg-card/60 p-4 font-mono text-sm">
<div className="text-muted-foreground select-none">
# Install all dependencies (Python venv, JS deps, etc.)
</div>
<div>just setup</div>
</div>
<div className="rounded-lg border border-border bg-card/60 p-4 font-mono text-sm">
<div className="text-muted-foreground select-none"># Build the app</div>
<div>just build</div>
</div>
</div>
<p className="mt-4 text-sm text-muted-foreground">
The built app will be in{' '}
<code className="text-xs bg-muted px-1.5 py-0.5 rounded">
tauri/src-tauri/target/release/bundle/
</code>
</p>
</div>
{/* Dev mode */}
<div>
<h2 className="text-sm font-medium text-muted-foreground uppercase tracking-wider mb-3">
Or run in dev mode
</h2>
<div className="rounded-lg border border-border bg-card/60 p-4 font-mono text-sm">
<div className="text-muted-foreground select-none">
# Start the dev server with hot reload
</div>
<div>just dev</div>
</div>
</div>
</div>
{/* Links */}
<div className="mt-12 pt-8 border-t border-border flex flex-wrap gap-4 text-sm">
<a
href={GITHUB_REPO}
target="_blank"
rel="noopener noreferrer"
className="text-muted-foreground hover:text-foreground transition-colors"
>
GitHub Repo
</a>
<a
href={`${GITHUB_REPO}/issues`}
target="_blank"
rel="noopener noreferrer"
className="text-muted-foreground hover:text-foreground transition-colors"
>
Report an issue
</a>
<a
href={`${GITHUB_REPO}/blob/main/CONTRIBUTING.md`}
target="_blank"
rel="noopener noreferrer"
className="text-muted-foreground hover:text-foreground transition-colors"
>
Contributing guide
</a>
</div>
</div>
</section>
<Footer />
</>
);
}
+2 -3
View File
@@ -294,14 +294,13 @@ export default function Home() {
{/* Linux */}
<a
href={downloadLinks.linux}
download
href="/linux-install"
className="flex items-center rounded-xl border border-border bg-card/60 backdrop-blur-sm px-5 py-4 transition-all hover:border-accent/30 hover:bg-card group"
>
<LinuxIcon className="h-6 w-6 shrink-0 text-muted-foreground group-hover:text-foreground transition-colors" />
<div className="ml-4">
<div className="text-sm font-medium">Linux</div>
<div className="text-xs text-muted-foreground">AppImage (x64)</div>
<div className="text-xs text-muted-foreground">Build from source</div>
</div>
</a>
</div>
@@ -20,7 +20,9 @@ export function unlockAudioContext() {
audioUnlocked = true;
// Unlock WaveSurfer's internal audio element
if (sharedWaveSurfer) {
// Skip if already playing — the context is already unlocked and the
// play/pause/reset dance would destroy the active playback.
if (sharedWaveSurfer && !sharedWaveSurfer.isPlaying()) {
const media = sharedWaveSurfer.getMediaElement();
if (media) {
media.muted = true;
+3 -2
View File
@@ -18,7 +18,7 @@ const GITHUB_API_BASE = 'https://api.github.com';
// Cache for release info (in-memory cache, resets on server restart)
let cachedReleaseInfo: ReleaseInfo | null = null;
let cacheTimestamp: number = 0;
const CACHE_DURATION = 1000 * 60 * 10; // 10 minutes
const CACHE_DURATION = 1000 * 60 * 5; // 5 minutes
// Cache for star count
let cachedStarCount: number | null = null;
@@ -36,7 +36,7 @@ export async function getLatestRelease(): Promise<ReleaseInfo> {
try {
const response = await fetch(`${GITHUB_API_BASE}/repos/${GITHUB_REPO}/releases/latest`, {
next: { revalidate: 600 }, // Revalidate every 10 minutes
cache: 'no-store',
headers: {
Accept: 'application/vnd.github.v3+json',
},
@@ -123,6 +123,7 @@ async function getTotalDownloads(): Promise<number> {
const response = await fetch(
`${GITHUB_API_BASE}/repos/${GITHUB_REPO}/releases?per_page=100&page=${page}`,
{
cache: 'no-store',
headers: { Accept: 'application/vnd.github.v3+json' },
},
);
+1 -1
View File
@@ -1,6 +1,6 @@
{
"name": "voicebox",
"version": "0.2.2",
"version": "0.2.3",
"private": true,
"workspaces": [
"app",
+1 -1
View File
@@ -1,7 +1,7 @@
{
"name": "@voicebox/tauri",
"private": true,
"version": "0.2.2",
"version": "0.2.3",
"type": "module",
"scripts": {
"dev": "vite",
+1 -1
View File
@@ -1,6 +1,6 @@
[package]
name = "voicebox"
version = "0.2.2"
version = "0.2.3"
description = "A production-quality desktop app for Qwen3-TTS voice cloning and generation"
authors = ["you"]
license = ""
+2 -2
View File
@@ -1,7 +1,7 @@
{
"$schema": "https://schema.tauri.app/config/2",
"productName": "Voicebox",
"version": "0.2.2",
"version": "0.2.3",
"identifier": "sh.voicebox.app",
"build": {
"beforeDevCommand": "bun run dev",
@@ -12,7 +12,7 @@
"bundle": {
"active": true,
"targets": "all",
"createUpdaterArtifacts": false,
"createUpdaterArtifacts": "v1Compatible",
"externalBin": ["binaries/voicebox-server"],
"icon": [
"icons/32x32.png",
+5 -1
View File
@@ -64,13 +64,17 @@ class TauriUpdater implements PlatformUpdater {
}
this.notifySubscribers();
} catch (error) {
const message = error instanceof Error ? error.message : String(error);
// Tauri updater throws on 404 / no published release / network errors.
// Treat "no update available" style errors as up-to-date, not failures.
const isNoUpdate = /404|not found|no update|up.to.date/i.test(message);
this.status = {
checking: false,
available: false,
downloading: false,
installing: false,
readyToInstall: false,
error: error instanceof Error ? error.message : 'Failed to check for updates',
error: isNoUpdate ? undefined : message,
};
this.notifySubscribers();
}
+1 -1
View File
@@ -1,7 +1,7 @@
{
"name": "@voicebox/web",
"private": true,
"version": "0.2.2",
"version": "0.2.3",
"type": "module",
"scripts": {
"dev": "vite",