claude-voice 1.0.0

package/scripts/listen.py

@@ -0,0 +1,161 @@
+#!/usr/bin/env python3
+"""
+Voice Input for Claude - Simple and Reliable
+Records audio, transcribes with OpenAI Whisper, copies to clipboard.
+
+Usage: python3 listen.py [seconds]
+Default: 5 seconds recording
+"""
+
+import os
+import sys
+import subprocess
+import tempfile
+import json
+import urllib.request
+
+def record_with_sounddevice(filename, duration, sample_rate=16000):
+    """Record using sounddevice (pip install sounddevice soundfile)"""
+    try:
+        import sounddevice as sd
+        import soundfile as sf
+    except ImportError:
+        return False
+
+    print(f"🎤 Recording for {duration} seconds... Speak now!")
+
+    audio = sd.rec(int(duration * sample_rate), samplerate=sample_rate, channels=1, dtype='int16')
+    sd.wait()
+
+    sf.write(filename, audio, sample_rate)
+    return True
+
+def record_with_pyaudio(filename, duration, sample_rate=16000):
+    """Record using PyAudio"""
+    try:
+        import pyaudio
+        import wave
+    except ImportError:
+        return False
+
+    CHUNK = 1024
+    FORMAT = pyaudio.paInt16
+    CHANNELS = 1
+
+    p = pyaudio.PyAudio()
+    stream = p.open(format=FORMAT, channels=CHANNELS, rate=sample_rate,
+                    input=True, frames_per_buffer=CHUNK)
+
+    print(f"🎤 Recording for {duration} seconds... Speak now!")
+
+    frames = []
+    for _ in range(0, int(sample_rate / CHUNK * duration)):
+        data = stream.read(CHUNK, exception_on_overflow=False)
+        frames.append(data)
+
+    stream.stop_stream()
+    stream.close()
+    p.terminate()
+
+    wf = wave.open(filename, 'wb')
+    wf.setnchannels(CHANNELS)
+    wf.setsampwidth(p.get_sample_size(FORMAT))
+    wf.setframerate(sample_rate)
+    wf.writeframes(b''.join(frames))
+    wf.close()
+    return True
+
+def transcribe_openai(audio_path):
+    """Transcribe using OpenAI Whisper API with curl"""
+    api_key = os.environ.get('OPENAI_API_KEY')
+    if not api_key:
+        print("❌ OPENAI_API_KEY not set")
+        return None
+
+    result = subprocess.run([
+        'curl', '-s',
+        'https://api.openai.com/v1/audio/transcriptions',
+        '-H', f'Authorization: Bearer {api_key}',
+        '-F', f'file=@{audio_path}',
+        '-F', 'model=whisper-1'
+    ], capture_output=True, text=True)
+
+    try:
+        response = json.loads(result.stdout)
+        return response.get('text', '')
+    except:
+        print(f"❌ API Error: {result.stdout}")
+        return None
+
+def copy_to_clipboard(text):
+    """Copy text to macOS clipboard"""
+    subprocess.run(['pbcopy'], input=text.encode(), check=True)
+
+def type_to_terminal(text):
+    """Auto-type text into the active terminal using AppleScript"""
+    # Escape special characters for AppleScript
+    escaped = text.replace('\\', '\\\\').replace('"', '\\"').replace('\n', '\\n')
+
+    script = f'''
+    tell application "System Events"
+        keystroke "{escaped}"
+        key code 36
+    end tell
+    '''
+
+    subprocess.run(['osascript', '-e', script], check=True)
+
+def main():
+    duration = int(sys.argv[1]) if len(sys.argv) > 1 else 5
+
+    print("╔════════════════════════════════════════╗")
+    print("║       Claude Voice Input               ║")
+    print("╚════════════════════════════════════════╝")
+    print("")
+
+    # Check API key
+    if not os.environ.get('OPENAI_API_KEY'):
+        print("❌ OPENAI_API_KEY not set")
+        print("   Run: export OPENAI_API_KEY='your-key'")
+        sys.exit(1)
+
+    # Create temp file
+    with tempfile.NamedTemporaryFile(suffix='.wav', delete=False) as f:
+        temp_path = f.name
+
+    try:
+        # Try recording methods
+        recorded = record_with_sounddevice(temp_path, duration)
+
+        if not recorded:
+            recorded = record_with_pyaudio(temp_path, duration)
+
+        if not recorded:
+            print("❌ No recording library available.")
+            print("   Install: pip3 install sounddevice soundfile")
+            sys.exit(1)
+
+        print("⏳ Transcribing with OpenAI Whisper...")
+
+        text = transcribe_openai(temp_path)
+
+        if text:
+            print(f"\n📝 \"{text}\"\n")
+
+            # Auto-type to terminal
+            try:
+                type_to_terminal(text)
+                print("✅ Typed into terminal!")
+            except:
+                # Fallback to clipboard
+                copy_to_clipboard(text)
+                print("✅ Copied to clipboard! Paste with Cmd+V")
+        else:
+            print("❌ No speech detected")
+
+    finally:
+        if os.path.exists(temp_path):
+            os.unlink(temp_path)
+
+if __name__ == "__main__":
+    main()

package/scripts/postinstall.js

@@ -0,0 +1,57 @@
+#!/usr/bin/env node
+/**
+ * Claude Voice Extension - Post-Install Setup
+ *
+ * This script runs after npm install to set up the extension.
+ */
+
+const fs = require('fs');
+const path = require('path');
+const os = require('os');
+
+const CONFIG_DIR = path.join(os.homedir(), '.claude-voice');
+const CONFIG_FILE = path.join(CONFIG_DIR, 'config.json');
+const DEFAULT_CONFIG = path.join(__dirname, '..', 'config', 'default.json');
+
+console.log('\n  Claude Voice Extension - Post-Install Setup\n');
+
+// 1. Create config directory
+if (!fs.existsSync(CONFIG_DIR)) {
+  fs.mkdirSync(CONFIG_DIR, { recursive: true });
+  console.log('  Created config directory:', CONFIG_DIR);
+}
+
+// 2. Copy default config if none exists
+if (!fs.existsSync(CONFIG_FILE)) {
+  if (fs.existsSync(DEFAULT_CONFIG)) {
+    fs.copyFileSync(DEFAULT_CONFIG, CONFIG_FILE);
+    console.log('  Created default configuration');
+  }
+}
+
+// 3. Detect platform and show relevant info
+const platform = os.platform();
+console.log(`\n  Platform: ${platform}`);
+
+if (platform === 'darwin') {
+  console.log('  TTS: macOS "say" command available (built-in)');
+  console.log('  Terminal: AppleScript injection available');
+} else if (platform === 'linux') {
+  console.log('  TTS: Install espeak for local TTS: sudo apt install espeak');
+  console.log('  Terminal: Install xdotool for input: sudo apt install xdotool');
+}
+
+// 4. Show optional dependencies
+console.log('\n  Optional Features:');
+console.log('  - Wake word detection: Requires PICOVOICE_ACCESS_KEY');
+console.log('    Get a free key at: https://picovoice.ai/');
+console.log('  - OpenAI TTS/STT: Requires OPENAI_API_KEY');
+console.log('  - ElevenLabs TTS: Requires ELEVENLABS_API_KEY');
+
+// 5. Next steps
+console.log('\n  Next Steps:');
+console.log('  1. Run interactive setup:  claude-voice setup');
+console.log('  2. Or start directly:      claude-voice start');
+console.log('  3. Check status:           claude-voice status');
+console.log('  4. Diagnose issues:        claude-voice doctor');
+console.log('');

package/scripts/record.sh

@@ -0,0 +1,79 @@
+#!/bin/bash
+#
+# Simple Voice Input for Claude (no dependencies)
+# Uses macOS QuickTime for recording
+#
+
+TEMP_FILE="/tmp/claude-voice-$$.m4a"
+DURATION=${1:-5}
+
+echo "╔════════════════════════════════════════╗"
+echo "║       Claude Voice Input               ║"
+echo "╚════════════════════════════════════════╝"
+echo ""
+
+if [ -z "$OPENAI_API_KEY" ]; then
+    echo "❌ OPENAI_API_KEY not set"
+    exit 1
+fi
+
+echo "🎤 Recording for ${DURATION} seconds... Speak now!"
+
+# Use afrecord (built into macOS) or screencapture for audio
+# Alternative: use say -i for interactive input
+osascript -e "
+    set tempFile to \"$TEMP_FILE\"
+    set duration to $DURATION
+
+    tell application \"QuickTime Player\"
+        activate
+        set newRecording to new audio recording
+        delay 0.5
+        start newRecording
+        delay duration
+        stop newRecording
+        export document 1 in POSIX file tempFile using settings preset \"Audio Only\"
+        close document 1 saving no
+        quit
+    end tell
+" 2>/dev/null
+
+# Wait for file
+sleep 1
+
+if [ ! -f "$TEMP_FILE" ]; then
+    echo "❌ Recording failed. Trying alternative method..."
+
+    # Fallback: Use afplay/arecord if available
+    if command -v rec &> /dev/null; then
+        rec -r 16000 -c 1 "$TEMP_FILE" trim 0 $DURATION 2>/dev/null
+    else
+        echo "❌ No recording tool available."
+        echo "   Install ffmpeg: brew install ffmpeg"
+        echo "   Or install sox: brew install sox"
+        exit 1
+    fi
+fi
+
+echo "⏳ Transcribing..."
+
+# Call OpenAI Whisper API
+RESPONSE=$(curl -s https://api.openai.com/v1/audio/transcriptions \
+    -H "Authorization: Bearer $OPENAI_API_KEY" \
+    -F "file=@$TEMP_FILE" \
+    -F "model=whisper-1")
+
+TEXT=$(echo "$RESPONSE" | python3 -c "import sys, json; print(json.load(sys.stdin).get('text', ''))" 2>/dev/null)
+
+rm -f "$TEMP_FILE"
+
+if [ -z "$TEXT" ]; then
+    echo "❌ Transcription failed"
+    exit 1
+fi
+
+echo ""
+echo "📝 \"$TEXT\""
+echo ""
+echo -n "$TEXT" | pbcopy
+echo "✅ Copied to clipboard!"

package/scripts/setup-hooks.sh

@@ -0,0 +1,22 @@
+#!/bin/bash
+#
+# Quick script to install/uninstall Claude Code hooks
+#
+
+SCRIPT_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )"
+PROJECT_DIR="$( cd "$SCRIPT_DIR/.." && pwd )"
+
+case "$1" in
+    install)
+        echo "Installing hooks..."
+        node "$PROJECT_DIR/dist/cli.js" install-hooks
+        ;;
+    uninstall)
+        echo "Uninstalling hooks..."
+        node "$PROJECT_DIR/dist/cli.js" uninstall-hooks
+        ;;
+    *)
+        echo "Usage: $0 {install|uninstall}"
+        exit 1
+        ;;
+esac

package/scripts/voice-input.sh

@@ -0,0 +1,66 @@
+#!/bin/bash
+#
+# Simple Voice Input for Claude
+# Records audio, transcribes with OpenAI Whisper, copies to clipboard
+#
+
+TEMP_FILE="/tmp/claude-voice-$$.wav"
+DURATION=${1:-5}  # Default 5 seconds, or pass as argument
+
+echo "╔════════════════════════════════════════╗"
+echo "║       Claude Voice Input               ║"
+echo "╚════════════════════════════════════════╝"
+echo ""
+
+# Check for OPENAI_API_KEY
+if [ -z "$OPENAI_API_KEY" ]; then
+    echo "❌ OPENAI_API_KEY not set"
+    exit 1
+fi
+
+# Check for ffmpeg (can install with: brew install ffmpeg)
+if ! command -v ffmpeg &> /dev/null; then
+    echo "❌ ffmpeg not found. Install with: brew install ffmpeg"
+    exit 1
+fi
+
+echo "🎤 Recording for ${DURATION} seconds..."
+echo "   (Speak now!)"
+echo ""
+
+# Record using ffmpeg with macOS audio input
+ffmpeg -f avfoundation -i ":0" -t "$DURATION" -ar 16000 -ac 1 -y "$TEMP_FILE" 2>/dev/null
+
+if [ ! -f "$TEMP_FILE" ]; then
+    echo "❌ Recording failed"
+    exit 1
+fi
+
+echo "⏳ Transcribing with OpenAI Whisper..."
+
+# Call OpenAI Whisper API
+RESPONSE=$(curl -s https://api.openai.com/v1/audio/transcriptions \
+    -H "Authorization: Bearer $OPENAI_API_KEY" \
+    -F "file=@$TEMP_FILE" \
+    -F "model=whisper-1")
+
+# Extract text from response
+TEXT=$(echo "$RESPONSE" | python3 -c "import sys, json; print(json.load(sys.stdin).get('text', ''))" 2>/dev/null)
+
+# Cleanup
+rm -f "$TEMP_FILE"
+
+if [ -z "$TEXT" ]; then
+    echo "❌ No speech detected or transcription failed"
+    echo "Response: $RESPONSE"
+    exit 1
+fi
+
+echo ""
+echo "📝 Transcript:"
+echo "   \"$TEXT\""
+echo ""
+
+# Copy to clipboard
+echo -n "$TEXT" | pbcopy
+echo "✅ Copied to clipboard! Paste with Cmd+V in Claude"