chore: add future feature requests

Author: DavidHruby1

docs/FEATURES.md

@@ -0,0 +1,80 @@
+# Future Features
+
+## 1. Custom Vocabulary / Personal Dictionary
+
+Let users define names, acronyms, project terms, product names, usernames, and domain-specific jargon.
+
+Use this vocabulary in the LLM rewrite prompt, and in STT prompts if the configured provider supports prompting.
+
+## 2. App-Specific Rewrite Prompts
+
+Detect the active foreground app or window and apply a matching rewrite prompt.
+
+Examples:
+
+- Slack: casual and concise.
+- Email: polished and professional.
+- Cursor/OpenCode: precise technical prompt.
+- Notes: clean bullets or paragraphs.
+
+## 3. Rewrite Modes / Styles
+
+Add selectable rewrite modes for common output shapes.
+
+Examples:
+
+- Raw transcription.
+- Clean dictation.
+- Professional email.
+- Casual message.
+- Bullet notes.
+- Coding prompt.
+- Translate to English.
+
+## 4. Snippets / Voice Shortcuts
+
+Allow spoken cues to expand into saved text snippets.
+
+Examples:
+
+- "insert calendar link" -> saved scheduling link.
+- "support signoff" -> saved support closing text.
+- "meeting intro" -> reusable meeting template.
+
+## 5. Clipboard Mode and Output Controls
+
+Support output modes beyond direct typing.
+
+Modes:
+
+- Type into active app.
+- Copy to clipboard.
+- Copy and type.
+
+This helps when focus changes or when an app does not handle `SendInput` well.
+
+## 6. Transcription History
+
+Keep a local history of recent transcriptions.
+
+Store timestamp, final text, target app/window if available, and pipeline warnings.
+
+Audio should stay off by default unless explicitly added later.
+
+## 7. Local Offline STT Backend
+
+Support local/private transcription through local OpenAI-compatible backends.
+
+Possible integrations:
+
+- `faster-whisper-server`
+- `whisper.cpp`
+- Other local Whisper-compatible servers
+
+First version can be provider presets/documentation rather than bundling models.
+
+## 8. Usage Dashboard
+
+Low-priority polish feature.
+
+Show simple usage stats such as words dictated, estimated time saved, and top target apps.