Skip to content

v26.3.31#1682

Merged
ROBERT-MCDOWELL merged 745 commits intomainfrom
v26
Apr 1, 2026
Merged

v26.3.31#1682
ROBERT-MCDOWELL merged 745 commits intomainfrom
v26

Conversation

@ROBERT-MCDOWELL
Copy link
Copy Markdown
Collaborator

@ROBERT-MCDOWELL ROBERT-MCDOWELL commented Apr 1, 2026

  • Fix lib/classes/std_filter.py to add __getattr__ passthrough so filters don't break libraries expecting full TextIOBase
  • Fix duplicate sys.stderr overwrite in all TTS engine files (bark, fairseq, glowtts, xtts, yourtts, vits, tortoise, tacotron) — now correctly wraps both sys.stdout and sys.stderr

dependabot bot and others added 30 commits March 26, 2026 14:13
Bumps the pip group with 1 update in the /components/audiocraft directory: [requests](https://github.com/psf/requests).


Updates `requests` from 2.32.3 to 2.33.0
- [Release notes](https://github.com/psf/requests/releases)
- [Changelog](https://github.com/psf/requests/blob/main/HISTORY.md)
- [Commits](psf/requests@v2.32.3...v2.33.0)

---
updated-dependencies:
- dependency-name: requests
  dependency-version: 2.33.0
  dependency-type: direct:production
  dependency-group: pip
...

Signed-off-by: dependabot[bot] <[email protected]>
…/audiocraft/pip-aa7cb66ac2

Bump requests from 2.32.3 to 2.33.0 in /components/audiocraft in the pip group across 1 directory
Copilot AI review requested due to automatic review settings April 1, 2026 03:04
Copy link
Copy Markdown

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

This PR bumps the project to v26.3.31 and implements “Blocks/Chapters” preview plus improved headless resume behavior, while also aligning Docker/Compose/Podman launch paths around the ebook2audiobook.command entrypoint.

Changes:

  • Add block/chapter preview state and persistable block/sentence resume flow for headless conversions.
  • Rework container entrypoints/compose profiles (CPU/GPU) and update docs/notebooks for new run patterns.
  • Extend config/model defaults (e.g., IPA mapping/regex, default speaker wiring) and adjust dependency sets.

Reviewed changes

Copilot reviewed 44 out of 47 changed files in this pull request and generated 14 comments.

Show a summary per file
File Description
VERSION.txt Version bump to 26.3.31.
uninstall.sh Normalizes uninstall script reference.
requirements.txt Adds cryptography/py-cpuinfo/pandas; adjusts indic-nlp placement.
README.md Updates Podman Compose commands to use profiles and GPU service name.
pyproject.toml Mirrors dependency additions/reordering from requirements.
podman-compose.yml Introduces CPU/GPU profiles, new gpu service, switches entrypoint to command script, adds tmp volume.
Notebooks/kaggle-ebook2audiobook.ipynb Creates run/ dir and uses unbuffered python for logs.
Notebooks/colab_ebook2audiobook.ipynb Same as Kaggle notebook adjustments (dirs + unbuffered python).
lib/core.py Major refactor: block-based state, resume logic, per-block sentence audio layout, updated alerting/session reset behavior.
lib/conf.py Adds CLI option allowlist; reorganizes settings and environment setup.
lib/conf_models.py Adds IPA regex, default speaker, and per-engine default voice wiring.
lib/conf_lang.py Adds ipa_mapping rules table.
lib/classes/tts_manager.py Updates engine API to write to a sentence file path (instead of numeric index).
lib/classes/tts_engines/yourtts.py Adapts engine to new sentence file path API; switches to cleanup_memory().
lib/classes/tts_engines/xtts.py Adapts engine to new sentence file path API; switches to cleanup_memory().
lib/classes/tts_engines/vits.py Adapts engine to new sentence file path API; switches to cleanup_memory().
lib/classes/tts_engines/tortoise.py Adapts engine to new sentence file path API; switches to cleanup_memory().
lib/classes/tts_engines/tacotron.py Adapts engine to new sentence file path API; switches to cleanup_memory().
lib/classes/tts_engines/presets/yourtts_presets.py Uses engine default voice setting for presets.
lib/classes/tts_engines/presets/xtts_presets.py Uses engine default voice setting for presets.
lib/classes/tts_engines/presets/vits_presets.py Uses engine default voice setting for presets.
lib/classes/tts_engines/presets/tortoise_presets.py Uses engine default voice setting for presets.
lib/classes/tts_engines/presets/tacotron_presets.py Uses engine default voice setting for presets.
lib/classes/tts_engines/presets/glowtts_presets.py Uses engine default voice setting for presets.
lib/classes/tts_engines/presets/fairseq_presets.py Uses engine default voice setting for presets.
lib/classes/tts_engines/presets/bark_presets.py Uses engine default voice setting for presets.
lib/classes/tts_engines/glowtts.py Adapts engine to new sentence file path API; renames local phonemized var to part_ipa.
lib/classes/tts_engines/fairseq.py Adapts engine to new sentence file path API; tweaks (commented) silence timing.
lib/classes/tts_engines/common/utils.py Renames memory cleanup API, revises VTT generation to traverse per-block sentence dirs.
lib/classes/tts_engines/common/headers.py Exposes sys and std stream filter helpers to engines.
lib/classes/tts_engines/bark.py Adapts engine to new sentence file path API; switches to cleanup_memory().
lib/classes/std_filter.py Adds std stream wrapper classes for engines.
lib/classes/device_installer.py Adds CPU baseline detection and numpy pinning guard for older x86 CPUs.
lib/init.py Re-exports new config/model items (cli options, fernet items, IPA mapping/pattern).
ebook2audiobook.egg-info/requires.txt Mirrors dependency additions/reordering.
ebook2audiobook.command Adds CLI option validation, version flag, Docker/Podman detection improvements, and unbuffered python execution.
ebook2audiobook.cmd Adds CLI option validation, version flag, Docker/Podman detection improvements, and unbuffered python execution.
dockerfiles/HuggingfaceDockerfile Reworks HF Spaces image to single-stage build with script-driven dependency install and runtime user setup.
Dockerfile Switches to single-stage build/runtime image and updates entrypoint to command script; adds run/tmp volumes.
docker-compose.yml Updates entrypoint to command script; adds tmp volume; bumps APP_VERSION arg.
components/sitecustomize.py Improves import-hook robustness; extends patches to huggingface_hub and transformers compat.
components/audiocraft/requirements.txt Updates requests version pin.
app.py Uses cli_options, creates session IDs for headless runs, updates batch conversion loop, and updates help text examples.
.gitignore Broadens ignores for outputs/cache dirs with .gitkeep exceptions.
.dockerignore Updates docker ignore rules to exclude outputs/models/tmp/run and tools.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants