v1.5.1

Versions

Component	Version
TTS-WebUI	v1.5.1
PyTorch	2.11.0
xformers	0.0.35
CUDA	12.8.1
Python	3.11
Runpod base image	2.4.19
Application Manager	1.3.0

Upstream TTS-WebUI changes (v1.4.1 to v1.5.1)

v1.5.1

• fix: uv extension APIs
• fix: docker uv compatibility via proxy
• fix: automatic browser opens correct port behind proxy

v1.5.0

• feat: proxy-based architecture enabled by default (Gradio is now fronted by an internal tree proxy)
• feat: proxy-based OAI API for (uv) extensions
• chore: migrate FireRedTTS2, VoxCPM, MegaTTS3, DIA, Piper TTS, and OpenVoice V2 extensions to uv

Docker image changes

• Updated nginx upstream port to 7770 (TTS-WebUI now runs Gradio on 7767 with the tree proxy on 7770 in --docker mode)
• Forward X-Forwarded-Host to the backend so Gradio builds file URLs with the public hostname (fixes audio playback on Runpod and other cloud deployments)
• Updated the app manager stop script to kill ports 7767 and 7770 instead of 7860
• Dropped the transformers<5.0.0 pin (upstream resolved tortoise compatibility, tortoise is no longer in core)
• Removed orphan /TTS-WebUI/voices/tortoise mkdir
• Trimmed deprecated keys from config.json (enable_queue, prevent_thread_lock, show_tips, file_directories, favicon_path)

v1.4.1

Important

Versioning change: This image now aligns its version numbers with the upstream TTS-WebUI repository. The previous release of this Docker image was 5.1.3 (using an independent numbering scheme). Going forward, the image tag will always match the upstream TTS-WebUI tag.

Versions

Component	Version
TTS-WebUI	v1.4.1
PyTorch	2.11.0
xformers	0.0.35
CUDA	12.8.1
Python	3.11
Runpod base image	2.4.19
Application Manager	1.3.0

Upstream TTS-WebUI changes (v1.1.0 to v1.4.1)

v1.4.1

• fix: repair old extensions and migrate to uv

v1.4.0

• feat: improve extension catalog
• feat: speech to text category

v1.2.0

• feat: new launcher compatibility
• feat: venv based uv extensions

v1.1.0

• Upgraded PyTorch to 2.11.0

Docker image changes

• Pinned transformers>=4.43.0,<5.0.0 for tortoise-tts compatibility
• Removed bark_voice_clone extension
• Added tortoise voices directory (/TTS-WebUI/voices/tortoise)
• Consolidated version variables: RELEASE now serves as both the image tag and the upstream TTS-WebUI git checkout tag

5.1.3

• Bumped base image to version 2.4.9.
• Bumped TTS-WebUI to version v0.3.0.

5.1.1

• Bumped base image to version 2.4.8 to resolve issues with syncing venv to persistent storage.
• Added ffmpeg dependencies.
• Added application manager to start and stop the GUI after installing extensions.
• Fixed an issue with conflicting ports for the React application.

5.0.2

• Bumped base image to version 2.4.7 to resolve issues with syncing venv to persistent storage.

5.0.1

• Fixed mixed content issues when using a Cloudflare proxy for Runpod.

5.0.0

• Bumped CUDA to version 12.8.
• Bumped torch to version 2.7.0.
• Bumped xformers to version 0.0.30.
• Bumped to latest Runpod base image version.
• Bumped to latest version of the TTS-WebUI.
• Added support for persistent Python virtual environment (venv).
• Resolved issue with installation of extensions.

4.1.0

• Renamed to tts-webui-docker to align with the TTS-WebUI project.
• Updated tts-generation-webui repo, installation, startup scripts, etc to TTS-WebUI.
• Updated to latest TTS-WebUI commit.

4.0.0

• Bumped CUDA to version 12.4.
• Bumped base image to version 2.4.2.
• Bumped Python to version 3.11.12.
• Bumped torch to version 2.6.0.
• Bumped xformers to version 0.0.29.post3.
• Bumped NodeJS to version 22.

3.4.1

• Bumped base Docker image to version 2.2.3.
• Improved syncing.
• Bumped TTS generation web ui to latest commit.