AllTalk TTS

AllTalk version 2 BETA availability. (How to get version 2)

AllTalk v2 BETA is out/available for download. See this link here for the discussion, this link here to download it and this link here for screenshots.

AllTalk v2 significantly enhances v1, introducing new features while addressing many previous issues. Although still evolving, v2 offers a stable build and is the recommended version for most users as this is where update and development work is now focused.

AllTalk version 1 (Below)

AllTalk version 1 is an updated version of the Coqui_tts extension for Text Generation web UI. Features include:

Can be run as a standalone application or part of :
- Text-generation-webui link
- SillyTavern link
- KoboldCPP link
Simple setup utlilty Windows & Linux.
API Suite and 3rd Party support via JSON calls: Can be used with 3rd party applications via JSON calls.
Model Finetuning: Train the model specifically on a voice of your choosing for better reproduction.
Local/Custom models: Use any of the XTTSv2 models (API Local and XTTSv2 Local).
Bulk TTS Generator/Editor: Generate hours of TTS into one big file or have something read back to you demo.
DeepSpeed: A 2-3x performance boost generating TTS. Screenshot
Low VRAM mode: Great for people with small GPU memory or if your VRAM is filled by your LLM.
Custom Start-up Settings: Adjust your default start-up settings. Screenshot
Narrarator: Use different voices for main character and narration. Example Narration
Optional wav file maintenance: Configurable deletion of old output wav files. Screenshot
Documentation: Fully documented with a built in webpage. Screenshot
Clear Console output: Clear command line output for any warnings or issues.

🟦 Screenshots

Index

🟦 Screenshots
🟩 Installation
🟪 Updating & problems with updating
🔵🟢 DeepSpeed Installation (Windows & Linux)
🆘 Support Requests, Troubleshooting & Feature requests
🟨 Help with problems
⚫ Finetuning a model
⬜ AllTalk TTS Generator
🟠 API Suite and JSON-CURL
🔴 Future to-do list & Upcoming updates

🛠️ About this project & me

AllTalk is a labour of love that has been developed, supported and sustained in my personal free time. As a solo enthusiast (not a business or team) my resources are inherently limited. This project has been one of my passions, but I must balance it with other commitments.

To manage AllTalk sustainably, I prioritize support requests based on their overall impact and the number of users affected. I encourage you to utilize the comprehensive documentation and engage with the AllTalk community discussion area. These resources often provide immediate answers and foster a supportive user network.

Should your inquiry extend beyond the documentation, especially if it concerns a bug or feature request, I assure you I’ll offer my best support as my schedule permits. However, please be prepared for varying response times, reflective of the personal dedication I bring to AllTalk. Your understanding and patience in this regard are greatly appreciated.

It's important to note that I am not the developer of any TTS models utilized by AllTalk, nor do I claim to be an expert on them, including understanding all their nuances, issues, and quirks. For specific TTS model concerns, I’ve provided links to the original developers in the Help section for direct assistance.

Thank you for your continued support and understanding.

💖 Showing Your Support

If AllTalk has been helpful to you, consider showing your support through a donation on my Ko-fi page. Your support is greatly appreciated and helps ensure the continued development and improvement of AllTalk.

🟩 Quick Setup (Text-generation-webui & Standalone Installation)

AllTalk version 1 - Quick setup scripts are available for users on Windows 10/11 and Linux. Instructional videos for both setup processes are linked below.

Ensure that Git is installed on your system as it is required for cloning the repository. If you do not have Git installed, visit Git's official website to download and install it.
Windows users must install C++ development tools for Python to compile Python packages. Detailed information and a link to these tools can be found in the help section Windows & Python requirements for compiling packages.

<details> <summary>QUICK SETUP - Text-Generation-webui</summary> <br>

For a step-by-step video guide, click here.

To set up AllTalk within Text-generation-webui, follow either method:

Download AllTalk Setup:
- Via Terminal/Console (Recommended):
  - cd \text-generation-webui\extensions\
  - git clone https://github.com/erew123/alltalk_tts
- Via Releases Page (Cannot be automatically updated after install as its not linked to Github):
  - Download the latest alltalk_tts.zip from Releases and extract it to \text-generation-webui\extensions\alltalk_tts\.
Start Python Environment:
- In the text-generation-webui folder, start the environment with the appropriate command:
  - Windows: cmd_windows.bat
  - Linux: ./cmd_linux.sh<br><br>
  If you're unfamiliar with Python environments and wish to learn more, consider reviewing Understanding Python Environments Simplified in the Help section.
Run AllTalk Setup Script:
- Navigate to the AllTalk directory and execute the setup script:
  - cd extensions
  - cd alltalk_tts
  - Windows: atsetup.bat
  - Linux: ./atsetup.sh
Install Requirements:
- Follow the on-screen instructions to install the necessary requirements. It's recommended to test AllTalk's functionality before installing DeepSpeed.

Note: Always activate the Text-generation-webui Python environment before making any adjustments or using Fine-tuning. Additional instructions for Fine-tuning and DeepSpeed can be found within the setup utility and on this documentation page.

</details> <details> <summary>QUICK SETUP - Standalone Installation</summary> <br>

For a step-by-step video guide, click here.

To perform a Standalone installation of AllTalk:

Get AllTalk Setup:
- Via Terminal/Console (Recommended):
  - Navigate to your preferred directory: cd C:\myfiles\
  - Clone the AllTalk repository: git clone https://github.com/erew123/alltalk_tts
- Via Releases Page (Cannot be automatically updated after install as its not linked to Github):
  - Download alltalk_tts.zip from Releases and extract it to your chosen directory, for example, C:\myfiles\alltalk_tts\.
Start AllTalk Setup:
- Open a terminal/command prompt, move to the AllTalk directory, and run the setup script:
  - cd alltalk_tts
  - Windows: atsetup.bat
  - Linux: ./atsetup.sh
Follow the Setup Prompts:
- Select Standalone Installation and then Option 1 and follow any on-screen instructions to install the required files. DeepSpeed is automatically installed on Windows based system, but will only work on Nvidia GPU's. Linux based system users will have to follow the DeepSpeed installation instructions.

If you're unfamiliar with Python environments and wish to learn more, consider reviewing Understanding Python Environments Simplified in the Help section.

Important: Do not use spaces in your folder path (e.g. avoid /my folder-is-this/alltalk_tts-main) as this causes issues with Python & Conda.

</details>

Refer to 🟩 Other installation notes for further details, including information on additional voices, changing IP, character card notes etc.

If you wish to understand AllTalks start-up screen, please read Understanding the AllTalk start-up screen in the Help section.

🟩 Docker Builds and Google Colab's

While an AllTalk Docker build exists, it's important to note that this version is based on an earlier iteration of AllTalk and was set up by a third party. At some point, my goal is to deepen my understanding of Docker and its compatibility with AllTalk. This exploration may lead to significant updates to AllTalk to ensure a seamless Docker experience. However, as of now, the Docker build should be considered a BETA version and isn't directly supported by me.

As for Google Colab, there is partial compatibility with AllTalk, though with some quirks. I am currently investigating these issues and figuring out the necessary adjustments to enhance the integration. Until I can ensure a smooth experience, I won't be officially releasing any Google Colab implementations of AllTalk.

🟩 Manual Installation - As part of Text generation web UI (inc. macOSX)

<details> <summary>MANUAL INSTALLATION - Text-Generation-webui</summary>

Manual Installation for Text Generation Web UI

If you're using a Mac or prefer a manual installation for any other reason, please follow the steps below. This guide is compatible with the current release of Text Generation Web UI as of December 2023. Consider updating your installation if it's been a while, update instructions here.

For a visual guide on the installation process, watch this video.

Navigate to Text Generation Web UI Folder:
- Open a terminal window and move to your Text Generation Web UI directory with:
  - cd text-generation-webui
Activate Text Generation Web UI Python Environment:
- Start the appropriate Python environment for your OS using one of the following commands:
  - For Windows: cmd_windows.bat
  - For Linux: ./cmd_linux.sh
  - For macOS: cmd_macos.sh
  - For WSL: cmd_wsl.bat
- Loading the Text Generation Web UI's Python environment is crucial. If unsure about what a loaded Python environment should look like, refer to this image and video guide.
If you're unfamiliar with Python environments and wish to learn more, consider reviewing Understanding Python Environments Simplified in the Help section.
Move to Extensions Folder:
- cd extensions
Clone the AllTalk TTS Repository:
- git clone https://github.com/erew123/alltalk_tts
Navigate to the AllTalk TTS Folder:
- cd alltalk_tts
Install Required Dependencies:
- Install dependencies for your machine type:
  - For Windows: pip install -r system\requirements\requirements_textgen.txt
  - For Linux/Mac: pip install -r system/requirements/requirements_textgen.txt
Optional DeepSpeed Installation:

If you're using an Nvidia graphics card on Linux or Windows and wish to install DeepSpeed, follow the instructions here.
Recommendation: Start Text Generation Web UI and ensure AllTalk functions correctly before installing DeepSpeed.

Start Text Generation Web UI:

Return to the main Text Generation Web UI folder using cd .. (repeat as necessary).
- Start the appropriate Python environment for your OS using one of the following commands:
  - For Windows: start_windows.bat
  - For Linux: ./start_linux.sh
  - For macOS: start_macos.sh
  - For WSL: start_wsl.bat
Load the AllTalk extension in the Text Generation Web UI session tab.
For any updates to AllTalk or for tasks like Finetuning, always activate the Text Generation Web UI Python environment first.

Refer to 🟩 Other installation notes for further details, including information on additional voices, changing IP, character card notes etc.

</details>

🟩 Manual Installation - As a Standalone Application

<details> <summary>MANUAL INSTALLATION - Run AllTalk as a Standalone with Text-generation-webui</summary>

Running AllTalk as a Standalone Application alongside Text Generation Web UI

If you have AllTalk installed as an extension of Text Generation Web UI but wish to run it as a standalone application, follow these steps:

Activate Text Generation Web UI Python Environment:
- Use the appropriate command for your operating system to load the Python environment:
  - Windows: cmd_windows.bat
  - Linux: ./cmd_linux.sh
  - macOS: cmd_macos.sh
  - WSL: cmd_wsl.bat
Navigate to the AllTalk Directory:
- Move to the AllTalk folder with the following commands:
  - cd extensions
  - cd alltalk_tts
Start AllTalk:
- Run AllTalk with the command:
  - python script.py
There are no additional steps required to run AllTalk as a standalone application from this point.

</details> <details> <summary>MANUAL INSTALLATION - Custom Install of AllTalk</summary>

Custom Installation of AllTalk

Support for custom Python environments is limted. Please read Custom Python environments Limitations Notice below this section.

To run AllTalk as a standalone application with a custom Python environment, ensure you install AllTalk's requirements into the environment of your choice. The instructions provided are generalized due to the variety of potential Python environments.

Python Compatibility: The TTS engine requires Python 3.9.x to 3.11.x. AllTalk is tested with Python 3.11.x. See TTS Engine details.
Path Names: Avoid spaces in path names as this can cause issues.
Custom Python Environments: If encountering issues potentially related to a custom environment, consider testing AllTalk with the quick setup standalone method that builds its own environment.

Quick Overview of Python Environments

If you're unfamiliar with Python environments and wish to learn more, consider reviewing Understanding Python Environments Simplified in the Help section.