Description:
Enable users to interact with Lumina via voice, allowing speech-to-text input and AI responses read aloud via text-to-speech. This will improve accessibility, support hands-free usage, and create a more natural conversational experience.

Acceptance Criteria:

1. Integrate the Web Speech API (or a suitable polyfill) for speech-to-text user input in the chat UI.
2. Add a “microphone” button in the chat input area to start/stop recording, with a clear visual recording indicator.
3. Convert captured audio to text and populate the chat input before sending as a message.
4. Use the Web Speech Synthesis API for client-side TTS to read AI responses aloud automatically.
5. Provide playback controls (play/pause, volume) for spoken responses.
6. Add settings toggles so users can enable/disable voice input and/or TTS separately.
7. Ensure graceful fallback to text-only mode on browsers that don’t support the Speech APIs.

Tasks:

• Research cross-browser compatibility for Web Speech APIs and choose polyfills if necessary.
• Create a VoiceInput React component with mic button and recording state UI.
• Hook up speech recognition to feed transcribed text into the existing chat input.
• Build a TextToSpeech service using Web Speech Synthesis, with UI controls for playback.
• Extend user profile settings to include voice input and TTS toggles.
• Add Playwright tests (or mocks) to verify the voice recording and playback flows.
• Update the README and in-app tooltips to guide users on using voice features.

Estimated Effort: ~2 sprints