TalkingHead Releases

Fixes Windows npm package installation issues by correcting archive formats and extraction methods to ensure compatibility across platforms.

Mar 22, 2026

Major feature release adding OIDC Provider Support for various providers like Authentik, Google, Okta, and Microsoft Entra ID.

Mar 22, 2026

Adds directory direct links for Pro users, batch editing capabilities for custom props, and fixes various bugs related to OAuth secrets and pagination.

Mar 14, 2026

This is a small patch release that adds manual WireGuard endpoint management for machines, Compose `stop_grace_period` support, and various networking reliability improvements.

Mar 9, 2026

World Monitor updates include a drag-to-reorder World Clock panel, live news HLS support, RTL language support, and bug fixes for aviation, feeds, and UI.

Mar 1, 2026

This release introduces container health monitoring with automatic rollback during rolling deployments, Compose spec support for `healthcheck`, `devices`, and `ulimits`, and various reliability improvements.

Mar 1, 2026

Focuses on aviation data, live news, and map improvements including a native mobile experience and an Iran Attacks map layer.

Mar 1, 2026

This release delivers significant performance improvements and a smoother CLI experience, featuring faster startup times, optimized file collection, and reduced package size.

Feb 28, 2026

This release focuses on performance optimization through the implementation of edge caching infrastructure and UI enhancements, including a redesigned settings panel and new government security advisory features.

Feb 27, 2026

This release addresses a critical issue on Linux systems running Wayland, specifically fixing an EGL_BAD_ALLOC error encountered with NVIDIA GPUs by improving GPU detection logic.

Feb 26, 2026

This release focuses on GStreamer plugin management for the AppImage build, ensuring that bundled plugins do not conflict with host system plugins to prevent runtime errors.

Feb 26, 2026

This release focuses on stabilizing the agent system by improving error reporting, search reliability, and content filter bypass capabilities. Key updates include a migration to a new DuckDuckGo API, enhanced diagnostic tools for LLM failures, and a temporary switch to a more stable model configuration.

Jan 6, 2026

This release adds German and French lip-sync language modules contributed by StephanWald, along with a new isRaw option in speakAudio that skips all internal breaks. Bug fixes for stream interface are also included.

Dec 8, 2025

Nov 9, 2025

A major update focused on UI improvements, adding an eraser feature, internationalization support for multiple languages, and tutorial animations.

Sep 14, 2025

This release significantly improves streaming stability with AudioWorklet reuse, introduces an experimental avatar-only mode supporting multiple avatars, and adds various new APIs including update option, head.dispose method, and exposure of animations and userData. Also adds English pronunciation support and auto-generates missing RPM blend shapes.

Sep 3, 2025

Introduces a new ZipVoice engine for high-quality, fast, and small text-to-speech output with cloning capabilities.

Jul 21, 2025

Focuses on bug fixes and documentation updates, addressing environment variable typos, input width issues, and adding contributing guidelines.

Jun 23, 2025

A maintenance release focused on bug fixes, documentation improvements, and CI configuration synchronization.

May 30, 2025

This release introduces streaming support for audio and lip-sync, along with Draco compression support for model loading. Both features were contributed by external contributors.

May 29, 2025