Skip to main content

What to expect from v0.2.0 release of AutoShorts

· 3 min read
Shafil Alam
Shafil Alam
Software Developer and Creator

We, actually just me, are excited to announce the future release of v0.2.0 of AutoShorts. This release brings major changes to the project and will be a significant milestone in the project's development.

Note this release is still in development and the features mentioned here are subject to change.

Here are some of the changes you can expect in the upcoming release:

  • New Web UI: The new version will feature a web UI that will allow users to interact with the project without having to use the command line. The web UI will provide a more user-friendly interface for creating videos and managing projects. The web UI will be built using Next.js and Tailwind CSS. The backend will be provided by the existing AutoShorts API.

  • New server support: The new version will feature a server mode that will allow users to run the project as a server. This will allow users to create videos programmatically using the AutoShorts API. The server mode will be built using Express.js and will provide a RESTful API for creating videos. Run via npx autoshorts --server. The server reads environment variables .env for configuration, such as the port, resource path, and temporary directory.

  • Added image, voice, and subtitle customization support: The new version will support image, voice, and subtitle customization. This will allow users to customize the appearance of the images, voices, and subtitles in their videos.

  • Option to not include background music or video: The new version will feature an option to not include background music or video in the final video. This option is useful for those who do not want to include irrelevant background music or video in their videos.

  • API key read from environment variables: The new version will support reading API keys from environment variables. This will allow users to store their API keys securely and avoid hardcoding them in the project configuration. Use .env file to store your API keys.

  • AI Output is now structured: The new version will feature structured AI output.The AI script generation has been be reworked to give each video type its own prompt to build its data object.

  • Added video orientation support: The new version will support video orientation. Users can now specify whether they want their video to be vertical or horizontal.

  • Forked ffcreator and inkpaint: The new version will feature a forked version of ffcreator and inkpaint as internal dependencies in packages directory. This will allow us to make changes to these libraries without affecting the original repositories.

  • OpenAI API support: The new version will support the OpenAI API. This will allow users to use GPT-4o, GPT-4o-mini, o1, and other OpenAI models to generate video scripts. This also supports custom endpoints to use Groq AI, local Ollama API, and other OpenAI compatible APIs.

  • Google Gemini AI API support: Users can now use the Google Gemini AI API to generate video scripts. This will allow users to use the Google Gemini AI model such as Gemini-1.5-Flash and Gemini-1.5-Pro to generate video scripts.

  • Anthropic (Claude) API support: Users can now use the Anthropic (Claude) API to generate video scripts. This will allow users to use the Claude 3.5 Sonnet, Claude 3 Opus, Claude 3 Sonnet, and Claude 3 Haiku models to generate video scripts.