Skip to main content

​
VoiceAgentClient

The main class for interacting with the Lokutor Voice Agent.

​
Constructor

client = VoiceAgentClient(
    api_key: str,
    prompt: str,
    voice: VoiceStyle = VoiceStyle.F1,
    language: Language = Language.ENGLISH,
    server_url: str = None,
    on_transcription: Callable = None,
    on_response: Callable = None,
    on_error: Callable = None
)
Parameters:
ParameterTypeDescription
api_keystrRequired. Your Lokutor API Key.
promptstrRequired. The system prompt defining the AI’s persona.
voiceVoiceStyleOptional. Default is VoiceStyle.F1.
languageLanguageOptional. Default is Language.ENGLISH.
server_urlstrOptional. Custom WebSocket URL.
on_transcriptionCallableCallback for user speech transcriptions.
on_responseCallableCallback for AI text responses.
on_errorCallableCallback for error events.

​
Methods

​
start_conversation()

Starts the conversation loop, automatically handling microphone input and speaker output. Blocks until the session is ended or interrupted.

​
disconnect()

Gracefully closes the connection to the Lokutor servers.

​
TTSClient

Dedicated client for converting text to high-quality streaming audio.

​
Constructor

client = TTSClient(api_key: str, server_url: str = None)

​
Methods

​
synthesize(options)

Starts the text-to-speech synthesis process. Parameters:
ParameterTypeDescription
textstrRequired. The text to speak.
voiceVoiceStyleOptional.
languageLanguageOptional.
speedfloatOptional. Default is 1.05.
stepsintOptional. Synthesis quality (1-50). Default is 24.
playboolOptional. Whether to play audio automatically.
blockboolOptional. Whether to wait for playback to finish.

​
Enums

​
VoiceStyle

  • F1 to F5: Female voices.
  • M1 to M5: Male voices.

​
Language

  • ENGLISH
  • SPANISH
  • FRENCH
  • PORTUGUESE
  • KOREAN