orpheus-tts 0.1.2

Python SDK for Orpheus TTS - stream high-quality speech

Orpheus TTS Python SDK

Stream high-quality speech from the Orpheus TTS model.

Installation

pip install orpheus-tts

Quick Start

from orpheus_tts import OrpheusClient

client = OrpheusClient()

# Stream audio chunks
for chunk in client.stream("Hello world!", voice="antoine"):
    # chunk is raw PCM bytes (int16, 48kHz, mono)
    process_audio(chunk)

Usage

Basic Streaming

from orpheus_tts import OrpheusClient

client = OrpheusClient()

# Stream and save to file
with open("output.pcm", "wb") as f:
    for chunk in client.stream("Welcome to Orpheus TTS!", voice="antoine"):
        f.write(chunk)

Save as WAV

import wave
from orpheus_tts import OrpheusClient

client = OrpheusClient()

# Get complete audio
audio_data = client.stream_to_bytes("Hello, this is a test.", voice="antoine")

# Save as WAV file
with wave.open("output.wav", "wb") as wav:
    wav.setnchannels(1)        # mono
    wav.setsampwidth(2)        # 16-bit
    wav.setframerate(48000)    # 48kHz
    wav.writeframes(audio_data)

Async Streaming

import asyncio
from orpheus_tts import OrpheusClient

async def main():
    client = OrpheusClient()
    
    async for chunk in client.stream_async("Async streaming!", voice="antoine"):
        await process_audio_async(chunk)

asyncio.run(main())

Real-time Playback with PyAudio

import pyaudio
from orpheus_tts import OrpheusClient

client = OrpheusClient()

# Initialize PyAudio
p = pyaudio.PyAudio()
stream = p.open(
    format=pyaudio.paInt16,
    channels=1,
    rate=48000,
    output=True
)

# Stream and play
for chunk in client.stream("This plays in real-time!", voice="antoine"):
    stream.write(chunk)

stream.stop_stream()
stream.close()
p.terminate()

Available Voices

from orpheus_tts import VOICES

print(VOICES)

Configuration

client = OrpheusClient(
    max_tokens=3000,        # Maximum tokens to generate
    temperature=1.0,        # Sampling temperature
    repetition_penalty=1.1, # Repetition penalty
)

# Or override per-request
for chunk in client.stream(
    "Custom settings for this request.",
    voice="antoine",
    max_tokens=2000,
    temperature=0.9,
):
    process(chunk)

Audio Format

All audio is returned as raw PCM with the following format:

• Sample Rate: 48,000 Hz
• Bit Depth: 16-bit signed integer (int16)
• Channels: Mono (1 channel)

Error Handling

from orpheus_tts import OrpheusClient, OrpheusError, AuthenticationError

client = OrpheusClient()

try:
    for chunk in client.stream("Hello!", voice="antoine"):
        process(chunk)
except AuthenticationError as e:
    print(f"Authentication failed: {e}")
except OrpheusError as e:
    print(f"TTS error: {e}")

License

MIT