This is a modified version of obsidian-edge-tts to add read-clipboard-aloud and some AI-related features.

Obsidian TTS Read Anything Plugin

Why Fork Instead of PR ?

In my daily life, I frequently need to read mathematics textbooks that contain numerous LaTeX expressions. Text-to-speech systems often struggle with these mathematical notations. As someone with ADHD, having an audio component helps me maintain focus while reading. This is why I needed clipboard reading functionality - to read processed content that has been converted to a more TTS-friendly format and copied to the clipboard.

An issue was created to include clipboard reading as a feature, but the original maintainer felt it did not align with the project’s direction.

This fork exists to incorporate clipboard support and AI-based rewriting to read LaTeX or explain the code block.

Integrated clipboard reading functionality allows for easier integration with AI plugins like QuickAdd or Text Generator that can write to the clipboard.

Overview

The Obsidian TTS Read Anything Plugin is a community plugin for Obsidian that allows you to read texts that are difficult for traditional TTS to read directly, such as mathematical formulas or code snippets, using AI.

Features

• Read selected text or entire notes aloud
• Choose from a list of top voices or specify a custom voice
• Adjust playback speed of voice over
• Toggle optional notices for playback status
• Option to generate mp3 file and embed it directly in note
• Listen to voice samples before selecting a voice (via tts.travisvn.com)

Added Features

• Read text directly from the clipboard
• AI rewriting of LaTeX formulas before reading aloud

Installation

Install Obsidian plugin

1. Open Obsidian
2. Go to Settings → Community Plugins
3. Search for TTS Read Anything
4. Click Install and then Enable

Alternatively, you can manually download the latest release from GitHub Releases

Usage

• Open the note you want to read aloud (or select the text)
• Use the Read note aloud command from the command palette
• Use the read-clipboard-aloud command from the command palette to read clipboard directly
  • or Click the ribbon icon (if enabled)
  • or Playback button in the status bar — this both starts a narration and then allows you to pause or resume once it's started
  • or Right-click on a file and select Read note aloud
• ✨ New ✨ Right-click on a file and select Generate MP3 to save the narration to an mp3 and embed it in the note

Settings

To access the plugin settings:

1. Go to Settings → Community Plugins → TTS Read Anything.
2. Configure the following options:
   • Select voice: Choose from a list of top voices.
   • Custom voice: Manually enter a custom voice.
   • Playback speed: Adjust playback speed multiplier.
   • Show notices: Toggle notices for playback status and errors.
   • Show status bar button: Toggle playback button in status bar.
   • Generate MP3: Toggle settings related to the Generate MP3 menu option.
   • Voice Samples: Visit tts.travisvn.com to sample available voices.

AI-related settings

• Enable AI LaTeX Processing: Converts LaTeX expressions into plain text via an AI service
• OpenAI Base URL: Base endpoint for AI requests
• AI Platform API Key: API key for the AI provider
• AI Model: Specific model for text processing

License

This fork remains under the terms of GPLv3.

Special thanks to travisvn, the original project maintainer, for their contributions to Obsidian-Edge-TTS