Quick Start Guide

This guide will help you get started with fmus-vox quickly.

Audio Processing

Load and Save Audio

from fmus_vox import Audio

# Load from file
audio = Audio.load("recording.wav")

# Save to file
audio.save("output.wav")

Record Audio

# Record for 5 seconds
audio = Audio.record(seconds=5)
audio.save("recording.wav")

Process Audio

# Chain operations
processed = (audio
            .normalize()
            .denoise()
            .resample(target_sr=16000))

Speech-to-Text

Simple Transcription

from fmus_vox import transcribe

text = transcribe("recording.wav")
print(f"Transcription: {text}")

With Specific Model

from fmus_vox import Transcriber

transcriber = Transcriber(model="whisper-large")
text = transcriber.transcribe("recording.wav", language="en")

Text-to-Speech

Simple Synthesis

from fmus_vox import speak

speak("Hello, world!", output="hello.wav")

With Voice Styling

from fmus_vox import Speaker

speaker = Speaker(voice="en-female-1")
speaker.set_style("happy").set_speed(1.2)
speaker.speak("This is exciting!", output="styled.wav")

Voice Cloning

Basic Cloning

from fmus_vox import clone_voice

clone_voice(
    reference_audio="my_voice.wav",
    text="Hello with my voice",
    output="cloned.wav"
)

Advanced Usage

from fmus_vox import VoiceCloner

cloner = VoiceCloner()
voice_id = cloner.add_reference("my_voice.wav")
audio = cloner.synthesize("Hello with my voice", voice_id)
audio.save("cloned.wav")

Wake Word Detection

from fmus_vox import WakeWordDetector

detector = WakeWordDetector(wake_word="hey assistant")
result = detector.detect("audio.wav")
if result:
    print("Wake word detected!")

Voice Application

from fmus_vox import VoiceApp

app = VoiceApp()

@app.on_wake("hey assistant")
def wake_handler():
    print("Wake word detected!")
    return True

@app.on_transcribe
def handle_transcription(text):
    print(f"User said: {text}")
    return process_command(text)

app.run()