Skip to content
Back to skills

Voice

ASecurity

Voice recognition, wake words, and voice-controlled trading

  • 2,496 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added February 10, 2026
blockchaintypescriptgoapi

Works with

  • api

Security analysis

A100/100

Pro scans all 2 files and shows the line behind each finding

Scanned February 12, 2026

npx -y skills add alsk1992/CloddsBot --skill voice --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Voice?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Voice
[![Security: A β€” Skills Directory](https://www.skillsdirectory.com/api/skills/alsk1992-voice/badge)](https://www.skillsdirectory.com/skills/alsk1992-voice)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: voice
description: "Voice recognition, wake words, and voice-controlled trading"
emoji: "🎀"
gates:
  envs:
    anyOf:
      - OPENAI_API_KEY
      - WHISPER_MODEL_PATH
---

# Voice - Complete API Reference

Voice-controlled interface with wake word detection, speech-to-text, and text-to-speech.

---

## Chat Commands

### Voice Control

```
/voice start                                Start voice listening
/voice stop                                 Stop voice listening
/voice status                               Check voice status
/voice wake "hey clodds"                    Set wake word
```

### Settings

```
/voice language en-US                       Set language
/voice sensitivity high                     Wake word sensitivity
/voice timeout 30                           Silence timeout (seconds)
/voice continuous on                        Enable continuous mode
```

---

## TypeScript API Reference

### Create Voice Assistant

```typescript
import { createVoiceAssistant } from 'clodds/voice';

const voice = createVoiceAssistant({
  // Wake word
  wakeWord: 'hey clodds',
  wakeWordSensitivity: 0.5,  // 0-1

  // Speech-to-text
  stt: {
    provider: 'whisper',  // 'whisper' | 'vosk' | 'google'
    model: 'base',        // 'tiny' | 'base' | 'small' | 'medium' | 'large'
    language: 'en',
  },

  // Text-to-speech
  tts: {
    provider: 'elevenlabs',  // 'elevenlabs' | 'say' | 'espeak'
    voice: 'rachel',
    speed: 1.0,
  },

  // Timeouts
  silenceTimeoutMs: 3000,
  maxRecordingMs: 30000,
});
```

### Start/Stop Listening

```typescript
// Start listening for wake word
await voice.start();

// Check if listening
const isListening = voice.isListening();

// Stop listening
await voice.stop();
```

### Event Handlers

```typescript
// Wake word detected
voice.on('wake', () => {
  console.log('Wake word detected!');
});

// Speech recognized
voice.on('speech', (text: string) => {
  console.log(`User said: ${text}`);
});

// Transcription complete
voice.on('transcript', (result) => {
  console.log(`Final: ${result.text}`);
  console.log(`Confidence: ${result.confidence}`);
  console.log(`Duration: ${result.durationMs}ms`);
});

// Error handling
voice.on('error', (error) => {
  console.error('Voice error:', error);
});

// Silence detected
voice.on('silence', () => {
  console.log('User stopped speaking');
});
```

### Speak (TTS)

```typescript
// Speak text
await voice.speak('Your order has been placed');

// Speak with options
await voice.speak('Market is up 5%', {
  voice: 'josh',
  speed: 1.2,
  pitch: 1.0,
});

// Cancel speaking
voice.cancelSpeech();
```

### Manual Transcription

```typescript
// Transcribe audio buffer
const result = await voice.transcribe(audioBuffer, {
  language: 'en',
  prompt: 'Trading commands',  // Context hint
});

console.log(`Text: ${result.text}`);
console.log(`Language: ${result.language}`);
```

### Voice Activity Detection

```typescript
// Check if user is speaking
const vad = voice.getVAD();

vad.on('speechStart', () => {
  console.log('User started speaking');
});

vad.on('speechEnd', (audio) => {
  console.log('User stopped speaking');
  // audio contains the recorded speech
});
```

---

## Wake Word Configuration

| Setting | Values | Description |
|---------|--------|-------------|
| `wakeWord` | Any phrase | Trigger phrase |
| `sensitivity` | 0.0 - 1.0 | Detection threshold |

**Built-in wake words:**
- "hey clodds"
- "okay clodds"
- "clodds"

---

## STT Providers

| Provider | Quality | Speed | Offline |
|----------|---------|-------|---------|
| **Whisper** | Excellent | Medium | Yes |
| **Vosk** | Good | Fast | Yes |
| **Google** | Excellent | Fast | No |

---

## TTS Providers

| Provider | Quality | Voices | API Key |
|----------|---------|--------|---------|
| **ElevenLabs** | Premium | 30+ | Required |
| **say** (macOS) | Good | System | None |
| **espeak** | Basic | Many | None |

---

## Voice Commands Examples

Once voice is active, speak naturally:

```
"Hey Clodds, what's my portfolio value?"
"Hey Clodds, buy 100 dollars of Trump YES"
"Hey Clodds, what are the top arbitrage opportunities?"
"Hey Clodds, set a price alert for Bitcoin at 100k"
```

---

## Best Practices

1. **Quiet environment** β€” Background noise affects accuracy
2. **Clear speech** β€” Speak clearly for best recognition
3. **Short commands** β€” Keep voice commands concise
4. **Confirmation** β€” Enable confirmations for trades
5. **Fallback** β€” Text input always available

Files in this skill

  • SKILL.md4.4 KB
  • index.ts11.8 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.