Skip to main content

Voice Integration

Voice integration components for building voice-enabled RecoAgent applications.

Overview​

The voice integration system provides comprehensive voice capabilities including speech-to-text, text-to-speech, and voice command processing.

Core Features​

  • Speech-to-Text: Convert speech to text
  • Text-to-Speech: Convert text to speech
  • Voice Commands: Process voice commands
  • Real-time Processing: Real-time voice processing
  • Multi-language Support: Support for multiple languages

Usage Examples​

Basic Voice Integration​

from recoagent.voice.integration import VoiceIntegration

# Create voice integration
voice_integration = VoiceIntegration(
    stt_provider="openai",
    tts_provider="elevenlabs"
)

# Process voice input
text_result = voice_integration.speech_to_text(
    audio_file="voice_input.wav",
    language="en-US"
)

print(f"Transcribed text: {text_result.text}")

Advanced Voice Processing​

from recoagent.voice.integration import AdvancedVoiceIntegration

# Create advanced voice integration
advanced_voice = AdvancedVoiceIntegration(
    stt_config={
        "provider": "openai",
        "model": "whisper-1",
        "language": "en"
    },
    tts_config={
        "provider": "elevenlabs",
        "voice": "alloy",
        "speed": 1.0
    }
)

# Process voice command
command_result = advanced_voice.process_voice_command(
    audio_file="command.wav",
    command_types=["search", "navigate", "execute"]
)

if command_result.recognized:
    print(f"Command: {command_result.command}")
    print(f"Parameters: {command_result.parameters}")

API Reference​

VoiceIntegration Methods​

speech_to_text(audio_file: str, language: str = "en-US") -> STTResult​

Convert speech to text

Parameters:

  • audio_file (str): Path to audio file
  • language (str): Language code

Returns: Speech-to-text result

text_to_speech(text: str, voice: str = "default") -> TTSResult​

Convert text to speech

Parameters:

  • text (str): Text to convert
  • voice (str): Voice to use

Returns: Text-to-speech result

See Also​