2:I[7012,["4765","static/chunks/4765-f5afdf8061f456f3.js","9856","static/chunks/9856-3b185291364d9bef.js","6687","static/chunks/app/docs/%5B...slug%5D/page-e07536548216bee4.js"],"MarkdownRenderer"]
4:I[9856,["4765","static/chunks/4765-f5afdf8061f456f3.js","9856","static/chunks/9856-3b185291364d9bef.js","6687","static/chunks/app/docs/%5B...slug%5D/page-e07536548216bee4.js"],""]
5:I[4126,[],""]
7:I[9630,[],""]
8:I[4278,["9856","static/chunks/9856-3b185291364d9bef.js","8172","static/chunks/8172-b3a2d6fe4ae10d40.js","3185","static/chunks/app/layout-2814fa5d15b84fe4.js"],"HeadingProvider"]
9:I[1476,["9856","static/chunks/9856-3b185291364d9bef.js","8172","static/chunks/8172-b3a2d6fe4ae10d40.js","3185","static/chunks/app/layout-2814fa5d15b84fe4.js"],"Header"]
a:I[3167,["9856","static/chunks/9856-3b185291364d9bef.js","8172","static/chunks/8172-b3a2d6fe4ae10d40.js","3185","static/chunks/app/layout-2814fa5d15b84fe4.js"],"Sidebar"]
b:I[7409,["9856","static/chunks/9856-3b185291364d9bef.js","8172","static/chunks/8172-b3a2d6fe4ae10d40.js","3185","static/chunks/app/layout-2814fa5d15b84fe4.js"],"PageFrame"]
3:T44ea,
# What's New in Voice Mode v4.1

Voice Mode v4.1 is a major release that introduces a voice-first interface, advanced speech processing, healthcare integrations, and comprehensive Arabic/Quranic language support.

## Release Highlights

- **Voice-First Input Bar**: Redesigned interface optimized for speech interaction
- **Streaming Text Display**: Real-time response rendering with smooth animations
- **Speaker Diarization**: Multi-speaker detection and attribution
- **FHIR R4 Streaming**: Healthcare data integration with retry resilience
- **Adaptive Audio Quality**: Dynamic bitrate adjustment based on network conditions
- **Quranic Lexicon**: 662 terms including all 114 Surah names and Tajweed terminology

---

## Getting Started

### Using the Voice-First Input Bar

1. **Tap the microphone button** to start recording
2. **Speak naturally** - the system detects when you stop speaking
3. **Wait for transcription** - your speech appears as text
4. **View the response** - streamed in real-time with Arabic support

### Selecting a VAD Preset

Choose the preset that matches your environment:

| Environment  | Recommended Preset |
| ------------ | ------------------ |
| Quiet room   | Sensitive          |
| Home/Office  | Balanced (default) |
| Public/Noisy | Relaxed            |

To change: **Settings (⚙️) → Voice Settings → VAD Preset**

### Understanding the Indicators

**Quality Badge (📶):**

- 🟢 **High**: Excellent connection (128kbps)
- 🟡 **Medium**: Acceptable quality (64kbps)
- 🟠 **Low**: Degraded quality (32kbps)
- 🔴 **Minimal**: Very poor connection (16kbps)

**PHI Indicator (🛡️):**

- 🟢 Green: No sensitive health info detected
- 🟡 Yellow: Potential PHI - review recommended
- 🔴 Red: PHI detected - secure handling active

### Adjusting Thinking Feedback

Control what happens while the AI processes your request:

- **Tone**: Soft chime when processing starts/ends
- **Haptic**: Gentle vibration feedback (mobile)
- **Visual**: Pulsing indicator animation

To configure: **Settings (⚙️) → Thinking Feedback**

---

## New Features

### Voice-First Input Bar

The input interface has been completely redesigned to prioritize voice interaction:

```
┌─────────────────────────────────────────────────┐
│  [🎤 Tap to Speak]  │  [⌨️]  │  [⚙️]           │
│                                                 │
│  "Ask me anything about the Quran..."           │
└─────────────────────────────────────────────────┘
```

**Key Features:**

- Large, prominent microphone button for easy tap-to-speak
- Visual audio level indicator during recording
- Keyboard toggle for text input when needed
- Settings quick-access for VAD and voice preferences

**Usage:**

1. Tap the microphone button to start recording
2. Speak your question naturally
3. Release or tap again to send
4. Watch the real-time transcription appear

### Streaming Text Display

Responses now stream in real-time with smooth character-by-character rendering:

**Features:**

- Token-by-token streaming from the AI
- Smooth CSS animations for text appearance
- Markdown rendering with syntax highlighting
- Citation links with hover previews
- RTL support for Arabic text segments

**Technical Details:**

- Adaptive chunk sizing based on network latency
- Automatic scroll-to-bottom during streaming
- Graceful handling of connection interruptions

### Adaptive VAD Presets

Voice Activity Detection now includes three presets for different environments:

| Preset        | Sensitivity | Best For                                |
| ------------- | ----------- | --------------------------------------- |
| **Sensitive** | High        | Silent rooms, minimal background noise  |
| **Balanced**  | Medium      | Typical home/office environments        |
| **Relaxed**   | Low         | Public spaces, background conversations |

**Settings Panel:**

```
Voice Settings
├── VAD Preset: [Balanced ▼]
├── Auto-stop delay: 1.5s
└── Push-to-talk: [ ] Enable
```

### PHI Indicator & Routing

Healthcare-compliant PHI (Protected Health Information) handling:

**Visual Indicator:**

- Green shield icon: No PHI detected
- Yellow shield: Potential PHI, review recommended
- Red shield: PHI detected, secure handling active

**Routing Features:**

- Automatic detection of 18 HIPAA identifiers
- Secure channel routing for PHI content
- Audit logging for compliance
- User notification when PHI is detected

### RTL Support for Arabic

Full right-to-left support for Arabic content:

**Features:**

- Automatic language detection
- Bidirectional text rendering
- RTL-aware UI layout
- Arabic numeral support
- Proper text alignment in mixed content

**Toggle:**

```
Display Settings
└── Text Direction: [Auto ▼] / LTR / RTL
```

### Unified Memory Context

Cross-session context management for personalized interactions:

**Capabilities:**

- Session history persistence
- User preference memory
- Conversation threading
- Context-aware follow-ups
- Learning style adaptation

---

## Phase 3 Features

### Speaker Diarization

Multi-speaker detection and attribution for group conversations:

**Capabilities:**

- Up to 4 concurrent speakers
- Real-time speaker change detection
- Speaker embedding extraction
- Cross-session speaker re-identification
- Confidence scoring per segment

**Use Cases:**

- Study circles with multiple participants
- Teacher-student Q&A sessions
- Family Quran recitation sessions

**Technical Specs:**

- Latency: <200ms for speaker change detection
- Accuracy: >90% speaker attribution
- Models: pyannote.audio segmentation

### FHIR R4 Streaming

Healthcare data integration with enterprise-grade resilience:

**Supported Resources:**

- Patient demographics
- Observations (vitals, lab results)
- Conditions (diagnoses)
- Medications
- Allergies

**Resilience Features:**

- Exponential backoff retry (1s → 2s → 4s → 8s)
- Circuit breaker pattern
- Partial response handling
- Connection pooling
- Health check monitoring

**Configuration:**

```yaml
fhir:
  base_url: https://fhir.example.com/r4
  timeout_ms: 5000
  max_retries: 3
  circuit_breaker_threshold: 5
```

### Adaptive Quality Controller

Dynamic audio quality adjustment based on network conditions:

**Quality Tiers:**

| Tier    | Bitrate | Sample Rate | Use Case          |
| ------- | ------- | ----------- | ----------------- |
| High    | 128kbps | 48kHz       | Excellent network |
| Medium  | 64kbps  | 24kHz       | Standard network  |
| Low     | 32kbps  | 16kHz       | Poor network      |
| Minimal | 16kbps  | 8kHz        | Very poor network |

**Hysteresis Behavior:**

- Upgrade: 10 consecutive good measurements
- Downgrade: 3 consecutive poor measurements
- Prevents quality oscillation

**Metrics Tracked:**

- RTT (Round Trip Time)
- Packet loss percentage
- Jitter
- Available bandwidth

---

## Lexicon Expansion

### Quranic Arabic Lexicon (328 terms)

Complete pronunciation coverage for Quranic content:

**Surah Names (114):**
All Surah names with accurate Modern Standard Arabic IPA:

- الفاتحة → /ʔalfaːtiħa/
- البقرة → /ʔalbaqara/
- آل عمران → /ʔaːl ʕimraːn/
- ... (all 114 Surahs)

**Tajweed Terms (50+):**

- تجويد → /tadʒwiːd/
- إدغام → /ʔidɣaːm/
- إخفاء → /ʔixfaːʔ/
- قلقلة → /qalqala/
- غنة → /ɣunna/
- مد → /madd/

**Islamic Vocabulary (200+):**

- Common phrases (Bismillah, Alhamdulillah)
- Names of Allah
- Prophet names
- Ritual terminology
- Theological concepts

### English Transliteration Lexicon (334 terms)

Transliterated Surah names and Islamic terms for English TTS:

**Multiple Spelling Variants:**

- Tajweed / tajwid
- Qur'an / Quran
- Insha'Allah / Inshallah

**IPA Approximations:**
Closest English phonemes for Arabic sounds:

- Al-Fatihah → /æl fɑːtiːhɑː/
- Bismillah → /bɪsmɪl lɑː/

---

## Configuration Reference

### Voice Settings

```javascript
// settings-panel.js configuration
voiceSettings: {
  vadPreset: 'balanced',    // 'sensitive' | 'balanced' | 'relaxed'
  autoStopDelay: 1500,      // ms
  pushToTalk: false,
  language: 'ar',           // 'ar' | 'en' | 'auto'
  rtlMode: 'auto'           // 'auto' | 'ltr' | 'rtl'
}
```

### Feature Flags

```python
# Feature flags for v4.1
VOICE_V4_INPUT_BAR = True
VOICE_V4_STREAMING_TEXT = True
VOICE_V4_SPEAKER_DIARIZATION = True
VOICE_V4_FHIR_STREAMING = True
VOICE_V4_ADAPTIVE_QUALITY = True
VOICE_V4_PHI_ROUTING = True
```

---

## Migration Guide

### From v4.0 to v4.1

**Breaking Changes:**

- None - v4.1 is fully backward compatible

**Recommended Updates:**

1. Enable new feature flags in configuration
2. Update client to use streaming text display
3. Configure VAD presets for your environment
4. Test lexicon pronunciations for Quranic content

**New Dependencies:**

```
pyannote.audio>=3.0.0  # Speaker diarization
fhir.resources>=7.0.0  # FHIR R4 support
```

---

## Performance Metrics

### Latency Targets

| Operation                    | Target  | Measured |
| ---------------------------- | ------- | -------- |
| Voice input to transcription | <500ms  | 320ms    |
| Speaker change detection     | <200ms  | 180ms    |
| Text streaming first token   | <300ms  | 250ms    |
| FHIR resource fetch          | <1000ms | 650ms    |
| Quality tier switch          | <100ms  | 80ms     |

### Resource Usage

| Component           | Memory | CPU |
| ------------------- | ------ | --- |
| Speaker diarization | +150MB | +5% |
| FHIR client         | +20MB  | +2% |
| Adaptive quality    | +5MB   | +1% |
| Lexicon service     | +10MB  | <1% |

---

## Known Issues

1. **Speaker diarization accuracy**: May decrease with >4 simultaneous speakers
2. **FHIR timeout**: First request after idle may timeout (connection pool warming)
3. **RTL mixed content**: Complex bidirectional text may occasionally misalign

---

## Acknowledgments

Voice Mode v4.1 was developed with contributions from:

- Platform team for core infrastructure
- Healthcare team for FHIR integration
- Localization team for Arabic/RTL support
- Community contributors for lexicon expansion

---

## UI Components Guide

> **Screenshots**: For visual reference, see the annotated screenshots in `/docs/voice/screenshots/`:
>
> - `voice-input-bar-states.png` - VoiceFirstInputBar in idle, recording, processing states
> - `streaming-text-rtl.png` - StreamingTextDisplay with RTL Arabic content
> - `quality-badge-tiers.png` - QualityBadge showing all 4 quality levels
> - `phi-indicator-states.png` - PHI indicator (green/yellow/red states)
> - `vad-presets-panel.png` - VAD preset selection in settings panel

### VoiceFirstInputBar Component

**Location:** `/var/www/quran/js/components/VoiceFirstInputBar.js`

The primary interface for voice interaction in v4.1:

```
┌─────────────────────────────────────────────────────────────┐
│                                                             │
│    ┌──────────────────────────────────────────────────┐    │
│    │  "What would you like to learn about today?"      │    │
│    └──────────────────────────────────────────────────┘    │
│                                                             │
│              ╭─────────────────────────╮                   │
│              │     [  🎤  ]            │  ← Tap to Speak   │
│              │   Recording...          │                   │
│              ╰─────────────────────────╯                   │
│                                                             │
│         [ ⌨️ Text ]    [ ⚙️ Settings ]    [ ❓ Help ]      │
│                                                             │
└─────────────────────────────────────────────────────────────┘
```

**States:**

| State      | Visual                 | Behavior                    |
| ---------- | ---------------------- | --------------------------- |
| Idle       | Grey microphone        | Tap to start recording      |
| Recording  | Pulsing red + waveform | Real-time audio levels      |
| Processing | Spinning indicator     | Transcription in progress   |
| Error      | Red outline + message  | Retry or switch to keyboard |

**Props:**

```javascript
<VoiceFirstInputBar
  onTranscript={(text) => handleSubmit(text)}
  vadPreset="balanced"
  language="ar"
  placeholder="Ask about any Surah..."
  showKeyboardToggle={true}
/>
```

### StreamingTextDisplay Component

**Location:** `/var/www/quran/js/components/StreamingTextDisplay.js`

Renders AI responses with real-time streaming:

```
┌─────────────────────────────────────────────────────────────┐
│  🤖 Assistant                                    12:34 PM   │
├─────────────────────────────────────────────────────────────┤
│                                                             │
│  Surah Al-Fatihah (الفاتحة) is the opening chapter of     │
│  the Quran. It consists of seven verses and is recited     │
│  in every unit of prayer...                                │
│                                                             │
│  **Key Themes:**                                           │
│  • Praise of Allah (verses 1-4)                            │
│  • Request for guidance (verses 5-7)                       │
│  • The straight path (الصراط المستقيم)█                   │
│                                                    ↑cursor │
│  ─────────────────────────────────────────────────────────│
│  📖 Source: Tafsir Ibn Kathir, Vol 1, p.23  [View →]       │
│                                                             │
└─────────────────────────────────────────────────────────────┘
```

**Features:**

- Token-by-token rendering with cursor animation
- Markdown support (bold, lists, code blocks)
- RTL text detection and rendering
- Citation card expansion on click
- Copy/share buttons on completion

**Props:**

```javascript
<StreamingTextDisplay
  stream={responseStream}
  onComplete={() => setIsStreaming(false)}
  showCitations={true}
  enableRTL="auto"
  animationSpeed={30} // ms per character
/>
```

### QualityBadge Component

**Location:** `/var/www/quran/js/components/QualityBadge.js`

Displays current audio quality and network status:

```
Normal view:          Expanded on hover/tap:
┌─────────┐          ┌─────────────────────────┐
│ 📶 High │          │ 📶 High Quality         │
└─────────┘          │ ─────────────────────── │
                     │ Bitrate: 128 kbps       │
                     │ RTT: 45ms               │
                     │ Packet Loss: 0.1%       │
                     └─────────────────────────┘
```

**Quality Indicators:**

| Badge   | Color  | Meaning              |
| ------- | ------ | -------------------- |
| 📶 High | Green  | Excellent connection |
| 📶 Med  | Yellow | Acceptable quality   |
| 📶 Low  | Orange | Degraded quality     |
| 📶 Min  | Red    | Minimal quality mode |

**Props:**

```javascript
<QualityBadge quality={networkQuality} showDetails={true} onQualityChange={(tier) => logQualityEvent(tier)} />
```

### Component Integration Example

```javascript
// Main voice interface integration
function VoiceInterface() {
  const [streaming, setStreaming] = useState(false);
  const [quality, setQuality] = useState("high");

  return (
    <div className="voice-interface">
      <QualityBadge quality={quality} />

      <StreamingTextDisplay stream={responseStream} onComplete={() => setStreaming(false)} showCitations={true} />

      <VoiceFirstInputBar onTranscript={handleSubmit} disabled={streaming} vadPreset={settings.vadPreset} />
    </div>
  );
}
```

---

## Related Documentation

- [Voice Mode Architecture](./voice-mode-v4-overview.md)
- [Speaker Diarization Service](./speaker-diarization-service.md)
- [FHIR Streaming Service](./fhir-streaming-service.md)
- [Adaptive Quality Service](./adaptive-quality-service.md)
- [Lexicon Service Guide](./lexicon-service-guide.md)

---

**Release Date:** December 2024
**Version:** 4.1.0
**Status:** Production Ready
6:["slug","voice/whats-new-v4-1","c"]
0:["X7oMT3VrOffzp0qvbeOas",[[["",{"children":["docs",{"children":[["slug","voice/whats-new-v4-1","c"],{"children":["__PAGE__?{\"slug\":[\"voice\",\"whats-new-v4-1\"]}",{}]}]}]},"$undefined","$undefined",true],["",{"children":["docs",{"children":[["slug","voice/whats-new-v4-1","c"],{"children":["__PAGE__",{},[["$L1",["$","div",null,{"children":[["$","div",null,{"className":"mb-6 flex items-center justify-between gap-4","children":[["$","div",null,{"children":[["$","p",null,{"className":"text-sm text-gray-500 dark:text-gray-400","children":"Docs / Raw"}],["$","h1",null,{"className":"text-3xl font-bold text-gray-900 dark:text-white","children":"What's New in Voice Mode v4.1"}],["$","p",null,{"className":"text-sm text-gray-600 dark:text-gray-400","children":["Sourced from"," ",["$","code",null,{"className":"font-mono text-xs","children":["docs/","voice/whats-new-v4-1.md"]}]]}]]}],["$","a",null,{"href":"https://github.com/mohammednazmy/VoiceAssist/edit/main/docs/voice/whats-new-v4-1.md","target":"_blank","rel":"noreferrer","className":"inline-flex items-center gap-2 rounded-md border border-gray-200 dark:border-gray-700 px-3 py-1.5 text-sm text-gray-700 dark:text-gray-200 hover:border-primary-500 dark:hover:border-primary-400 hover:text-primary-700 dark:hover:text-primary-300","children":"Edit on GitHub"}]]}],["$","div",null,{"className":"rounded-lg border border-gray-200 dark:border-gray-800 bg-white dark:bg-gray-900 p-6","children":["$","$L2",null,{"content":"$3"}]}],["$","div",null,{"className":"mt-6 flex flex-wrap gap-2 text-sm","children":[["$","$L4",null,{"href":"/reference/all-docs","className":"inline-flex items-center gap-1 rounded-md bg-gray-100 px-3 py-1 text-gray-700 hover:bg-gray-200 dark:bg-gray-800 dark:text-gray-200 dark:hover:bg-gray-700","children":"← All documentation"}],["$","$L4",null,{"href":"/","className":"inline-flex items-center gap-1 rounded-md bg-gray-100 px-3 py-1 text-gray-700 hover:bg-gray-200 dark:bg-gray-800 dark:text-gray-200 dark:hover:bg-gray-700","children":"Home"}]]}]]}],null],null],null]},[null,["$","$L5",null,{"parallelRouterKey":"children","segmentPath":["children","docs","children","$6","children"],"error":"$undefined","errorStyles":"$undefined","errorScripts":"$undefined","template":["$","$L7",null,{}],"templateStyles":"$undefined","templateScripts":"$undefined","notFound":"$undefined","notFoundStyles":"$undefined"}]],null]},[null,["$","$L5",null,{"parallelRouterKey":"children","segmentPath":["children","docs","children"],"error":"$undefined","errorStyles":"$undefined","errorScripts":"$undefined","template":["$","$L7",null,{}],"templateStyles":"$undefined","templateScripts":"$undefined","notFound":"$undefined","notFoundStyles":"$undefined"}]],null]},[[[["$","link","0",{"rel":"stylesheet","href":"/_next/static/css/7f586cdbbaa33ff7.css","precedence":"next","crossOrigin":"$undefined"}]],["$","html",null,{"lang":"en","className":"h-full","children":["$","body",null,{"className":"__className_f367f3 h-full bg-white dark:bg-gray-900","children":[["$","a",null,{"href":"#main-content","className":"skip-to-content","children":"Skip to main content"}],["$","$L8",null,{"children":[["$","$L9",null,{}],["$","$La",null,{}],["$","main",null,{"id":"main-content","className":"lg:pl-64","role":"main","aria-label":"Documentation content","children":["$","$Lb",null,{"children":["$","$L5",null,{"parallelRouterKey":"children","segmentPath":["children"],"error":"$undefined","errorStyles":"$undefined","errorScripts":"$undefined","template":["$","$L7",null,{}],"templateStyles":"$undefined","templateScripts":"$undefined","notFound":[["$","title",null,{"children":"404: This page could not be found."}],["$","div",null,{"style":{"fontFamily":"system-ui,\"Segoe UI\",Roboto,Helvetica,Arial,sans-serif,\"Apple Color Emoji\",\"Segoe UI Emoji\"","height":"100vh","textAlign":"center","display":"flex","flexDirection":"column","alignItems":"center","justifyContent":"center"},"children":["$","div",null,{"children":[["$","style",null,{"dangerouslySetInnerHTML":{"__html":"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}"}}],["$","h1",null,{"className":"next-error-h1","style":{"display":"inline-block","margin":"0 20px 0 0","padding":"0 23px 0 0","fontSize":24,"fontWeight":500,"verticalAlign":"top","lineHeight":"49px"},"children":"404"}],["$","div",null,{"style":{"display":"inline-block"},"children":["$","h2",null,{"style":{"fontSize":14,"fontWeight":400,"lineHeight":"49px","margin":0},"children":"This page could not be found."}]}]]}]}]],"notFoundStyles":[]}]}]}]]}]]}]}]],null],null],["$Lc",null]]]]
c:[["$","meta","0",{"name":"viewport","content":"width=device-width, initial-scale=1"}],["$","meta","1",{"charSet":"utf-8"}],["$","title","2",{"children":"What's New in Voice Mode v4.1 | Docs | VoiceAssist Docs"}],["$","meta","3",{"name":"description","content":"Complete feature summary for Voice Mode v4.1 release"}],["$","meta","4",{"name":"keywords","content":"VoiceAssist,documentation,medical AI,voice assistant,healthcare,HIPAA,API"}],["$","meta","5",{"name":"robots","content":"index, follow"}],["$","meta","6",{"name":"googlebot","content":"index, follow"}],["$","link","7",{"rel":"canonical","href":"https://assistdocs.asimo.io"}],["$","meta","8",{"property":"og:title","content":"VoiceAssist Documentation"}],["$","meta","9",{"property":"og:description","content":"Comprehensive documentation for VoiceAssist - Enterprise Medical AI Assistant"}],["$","meta","10",{"property":"og:url","content":"https://assistdocs.asimo.io"}],["$","meta","11",{"property":"og:site_name","content":"VoiceAssist Docs"}],["$","meta","12",{"property":"og:type","content":"website"}],["$","meta","13",{"name":"twitter:card","content":"summary"}],["$","meta","14",{"name":"twitter:title","content":"VoiceAssist Documentation"}],["$","meta","15",{"name":"twitter:description","content":"Comprehensive documentation for VoiceAssist - Enterprise Medical AI Assistant"}],["$","meta","16",{"name":"next-size-adjust"}]]
1:null