Create a web application for organizing user tasks.

Each task should include the following fields:
- Title
- Description
- Deadline date (only year, month, day)
- Priority level
- Category

Users should be able to define custom categories and custom priority levels.
Default values for custom categories and custom priority levels can be set in a settings page.

Include a calendar view for visualizing tasks.

Tasks can be created in two ways:

- Manually, using a standard input form.

- By voice command — the user clicks a microphone button and describes the task verbally.

	- The system records the audio and uses a public Speech-to-Text API to convert the speech into text. Use 'whisper-large-v3' model.

	- The transcribed text is then sent to a public LLM API, which extracts and structures the task details (title, description, deadline, etc.). Use 'moonshotai/kimi-k2-instruct-0905' model.

	- The structured task is then saved into the database.
	
	- The deadline date, priority level, and category are not required.
	
	- If there is an error while parsing the structured data, display which task fields were successfully extracted and ask for confirmation. 

Tasks can be edited.

Implementation details:

- Build the app using a single HTML file with JavaScript for functionality and Twilight CSS for styling.

- Use the browser’s local storage as the database for storing tasks.


Public Speech-to-Text API usage example:
curl https://api.groq.com/openai/v1/audio/translations \
  -H "Authorization: bearer ${GROQ_API_KEY}" \
  -F "file=@./sample_audio.m4a" \
  -F model=whisper-large-v3 \
  -F prompt="Specify context or spelling" \
  -F language="en" \
  -F temperature=0 \
  -F response_format=json
  
Public LLM API usage example:
curl "https://api.groq.com/openai/v1/chat/completions" \
  -X POST \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer ${GROQ_API_KEY}" \
  -d '{
         "messages": [
           {
             "role": "user",
             "content": ""
           }
         ],
         "model": "moonshotai/kimi-k2-instruct-0905",
         "temperature": 0.6,
         "max_completion_tokens": 4096,
         "top_p": 1,
         "stream": true,
         "stop": null
       }'