[FEAT]: Implement OpenAI Whisper for audio-to-text transcription

[marcvergees]

📝 Description

Add OpenAI Whisper audio transcription capability to FireForm, enabling users to record audio input that is automatically converted to text. The feature should include a backend API endpoint for audio processing and a frontend micro button that triggers recording, with an editable textarea for post-conversion corrections.

💡 Rationale

Audio transcription improves accessibility and user experience for form filling, particularly for users who prefer voice input or have mobility constraints. Integrating OpenAI Whisper provides accurate, reliable speech-to-text conversion that can be corrected by users before form submission.

🛠️ Proposed Solution

Implement audio transcription functionality across backend and frontend:

Backend:

• Create new API endpoint to receive audio files
• Integrate OpenAI Whisper API for audio-to-text conversion
• Add audio file validation and error handling
• Update requirements.txt with OpenAI library dependency
• Add Whisper API key configuration to environment variables

Frontend:

• Create micro button component to trigger audio recording
• Implement audio recording functionality with browser's Web Audio API
• Add textarea component to display and allow editing of transcribed text
• Handle audio upload to backend and display loading state
• Display transcription results in textarea for user correction
• Integrate transcription button with existing form fields

✅ Acceptance Criteria

How will we know this is finished?

• Backend API endpoint accepts and processes audio files
• Whisper successfully converts audio to text with reasonable accuracy
• Frontend micro button records audio and uploads to backend
• Transcribed text displays in editable textarea
• Users can modify transcribed text before submission
• Feature works in Docker container
• Error handling for failed transcriptions or network issues
• Documentation updated in docs/ with Whisper setup and usage
• Audio recording works across modern browsers (Chrome, Firefox, Safari, Edge)

📌 Additional Context

• Consider audio format support (MP3, WAV, WebM, etc.)
• Implement timeout for maximum recording duration
• Add visual feedback during recording and transcription processing
• Consider privacy implications and data retention policies for audio files