# what is whisper plugin
The Whisper plugin for Obsidian, developed by Nikita Danilov, is a powerful tool that integrates OpenAI's Whisper API to provide speech-to-text functionality within the Obsidian note-taking app. This plugin allows users to convert their spoken words into written notes effortlessly, making it convenient to capture ideas, meetings, or any spoken content directly into Obsidian.
### Key Features:
1. **Speech-to-Text Conversion**: The plugin transcribes spoken words into text, creating a new note with the transcribed content and the linked audio file.
2. **Recording Controls**: Users can start, pause, and stop recordings using an intuitive interface or keyboard shortcuts (`Alt + Q`).
3. **Upload Existing Audio Files**: Users can transcribe pre-recorded audio files by uploading them through the command palette.
4. **API Integration**: Requires an OpenAI API key to access the Whisper service for transcription.
5. **Customization Options**: Users can specify settings such as the API endpoint, model choice, language, and whether to save the audio files alongside the transcriptions.
### Installation:
1. Install the plugin from the "Community Plugins" section within Obsidian.
2. Provide your OpenAI API key in the plugin settings.
### Usage:
- **Start Recording**: Click the ribbon button or use the shortcut to begin recording.
- **Transcribe Audio Files**: Upload an existing audio file via the command palette, and the plugin will transcribe it into a new note.
This plugin enhances the functionality of Obsidian by enabling hands-free note-taking and easy transcription of audio content, making it a valuable tool for users who prefer voice input over typing.
For more detailed information, you can visit the [GitHub repository](https://github.com/nikdanilov/whisper-obsidian-plugin).
# how to install
> [!ai]+ AI
>
> ## How to Install the Whisper Plugin in Obsidian
> Installing the Whisper plugin in Obsidian is a straightforward process. Follow these steps to get it up and running:
> ### Step-by-Step Installation Guide:
> 1. **Open Obsidian**:
> - Launch the Obsidian application on your computer.
> 2. **Access Community Plugins**:
> - Click on the "Settings" gear icon located in the bottom left corner of the Obsidian window.
> - In the settings menu, navigate to the "Community plugins" section.
> 3. **Browse Community Plugins**:
> - If this is your first time using community plugins, you may need to enable them by clicking "Turn on community plugins".
> - Click on "Browse" to open the community plugin browser.
> 4. **Search for Whisper Plugin**:
> - In the search bar at the top of the community plugin browser, type "Whisper".
> - Find "Whisper" by Nikita Danilov in the search results.
> 5. **Install and Enable Whisper Plugin**:
> - Click on the plugin name to open its details page.
> - Click on the "Install" button.
> - Once installed, click on "Enable" to activate the plugin within your Obsidian setup.
> 6. **Configure API Key**:
> - After enabling the plugin, return to the main settings menu.
> - Find and click on "Whisper Settings" under installed plugins.
> - Enter your OpenAI API key in the provided field. You can obtain an API key from OpenAI's website if you don't already have one.
> ### Using Whisper Plugin:
> - **Start Recording**:
> - You can start recording by clicking a ribbon button that appears after enabling Whisper or using a keyboard shortcut (`Alt + Q`).
> - **Transcribe Existing Audio Files**:
> - To transcribe pre-recorded audio files, open the command palette (usually `Ctrl + P` or `Cmd + P`) and select “Upload audio file for transcription”. Choose your file, and Whisper will create a new note with both transcription and linked audio.
> ### Note:
> Ensure that you have a working internet connection since transcriptions are processed via OpenAI's online service.
> By following these steps, you'll be able to integrate powerful speech-to-text functionality into your Obsidian note-taking workflow with ease using Whisper.
> For more detailed information or troubleshooting, visit [Whisper's GitHub repository](https://github.com/nikdanilov/whisper-obsidian-plugin).
with screenshots:
![[../../media/Pasted image 20240623001118.png]]
![[../../media/Pasted image 20240623013451.png]]
![[../../media/Pasted image 20240623013457.png]]
insert your openai api key:
![[../../media/Pasted image 20240623013515.png]]
and setup the plugin:
![[../../media/Pasted image 20240623013527.png]]
![[../../media/Pasted image 20240623013533.png]]
remove "en" to enable speech recognition in any language:
![[../../media/Pasted image 20240623013554.png]]
and setup hotkeys:
![[../../media/Pasted image 20240623013621.png]]
![[../../media/Pasted image 20240623013627.png]]