pix2surf Releases
129 releases of miraymen/pix2surf
- v0.16.2
This release focuses on bug fixes for environment variables, text input widths, and interrupt handling, alongside the addition of contributing guidelines.
Jun 23, 2025
- v1.3Audio2Face-3D Samples v1.3
Jun 16, 2025
- v0.16.1
This patch release addresses timeout error detection and default option application, updates documentation to use `$EDITOR`, and includes various lint and formatting fixes.
May 30, 2025
- v1.2Audio2Face-3D Samples v1.2
Apr 10, 2025
- v0.16.0
A feature-rich release that adds an item position counter to the help screen and fixes multiple bugs across choose, confirm, spin, and table components, including Fish shell completion.
Mar 11, 2025
- v1.1.2
This update adds ADB version information to the About dialog and fixes self-update functionality to handle Intel and Apple Silicon architectures correctly.
Mar 10, 2025
- v1.1.1
A critical bug fix release addressing a self-update issue, followed by optimizations for the Auto theme, ADB retry logic, and text shaping improvements for CJK characters.
Mar 10, 2025
- v0.15.2
Fixed a bug where memory and CPU usage spiked when using soft-wrap in the pager component.
Jan 28, 2025
- v0.0.27
Updated dependencies and added new username extraction techniques to the social media username extractor.
Dec 12, 2024
- 5.2.0
Added a template engine and enabled fixed positioning for headers and footers in the PDF generation tool.
Nov 11, 2024
- 5.1.6
Fixed issues with MVT field read-only states and CSS errors in the date picker component.
Oct 26, 2024
- v3.1.2GUI for precise control
Released a new GUI for precise control over utterance sounds, allowing pitch and duration manipulation across 7000+ languages.
Oct 7, 2024
- v3.1.1Improved Vocoder through end-to-end Training
Improved vocoder quality by training it on synthetic spectrograms instead of ground-truth audio, resulting in a major boost in audio quality.
Sep 22, 2024
- v3.1Improved TTS in 7000 Languages
Enhanced TTS capabilities in 7000 languages by introducing stochastic prosody prediction and expanded language support.
Jul 25, 2024
- 2.2.6v2.2.6
Bug fixes in the LLM SDK regarding API types and chain arguments, along with Opik UI improvements like audit mode and mobile onboarding.
Jul 17, 2024
- v0.15.0Sherlock v0.15.0
Major update introducing official packaging on PyPI, Docker, and Fedora, refactoring the project structure using Poetry and pytest, and deprecating Tor support.
Jul 8, 2024
- 2.2.5v2.2.5
Updated LLM SDK methods and added various Opik improvements including span size limits, OQL parser fixes, and mobile UI refactors.
Jun 21, 2024
- v3.0TTS in 7000 Languages
This release dramatically expands the toolkit's capabilities by supporting over 7000 languages, significantly improving audio quality, and introducing watermarking features to prevent misuse. It marks a significant milestone as the first model of its kind to achieve such comprehensive multilingual coverage.
Jun 10, 2024
- 2.pPrompting Controlled Emotional TTS
This release introduces prompting capabilities to control the emotional tone of synthesized speech, allowing users to condition models on emotional prompts during training and inference. It provides the necessary tools and training interfaces to fine-tune models for expressive voice generation.
Jun 10, 2024
- 2.2.4v2.2.4
Comet LLM v2.2.4 introduces tests against real backends and updates the logging endpoint for chains and prompts.
May 16, 2024