OuteAI Python SDK

Official Python SDK for the OuteAI text-to-speech API at https://outeai.com.

Installation

From PyPI:

pip install outeai

From this repository:

cd outeai
pip install .

Build distributable artifacts:

cd outeai
python -m pip install build
python -m build

Install async support:

pip install "outeai[async]"

Authentication

Create an API key in your OuteAI account, then pass it to the client with Authorization: Bearer <key>.

The SDK uses https://outeai.com by default, so you do not need to set a base URL unless you want to override it.

Quick Start

from outeai import OuteAI

client = OuteAI(api_key="YOUR_OUTEAI_API_KEY")

voices = client.list_voices()
voice_id = voices[0]["voice_id"]

audio = client.generate_speech(
    text="Hello from OuteAI.",
    voice_id=voice_id,
)

audio.save("hello.wav")
client.close()

Streaming Audio

from outeai import OuteAI

client = OuteAI(api_key="YOUR_OUTEAI_API_KEY")

with client.stream_speech(
    text="This is a streaming response from OuteAI.",
    voice_id="vc_your_voice_id",
) as stream:
    stream.save("streamed.wav")

You can also iterate the WAV bytes yourself:

with client.stream_speech(
    text="Chunked audio output.",
    voice_id="vc_your_voice_id",
) as stream:
    for chunk in stream:
        print(f"received {len(chunk)} bytes")

Voice IDs

Voice clone management is intentionally not part of this SDK surface. Pass an existing voice_id into the TTS methods.

Async Usage

import asyncio

from outeai import AsyncOuteAI


async def main() -> None:
    client = AsyncOuteAI(api_key="YOUR_OUTEAI_API_KEY")

    voices = await client.list_voices()
    voice_id = voices[0]["voice_id"]

    audio = await client.generate_speech(
        text="Hello from the async OuteAI client.",
        voice_id=voice_id,
    )
    audio.save("hello-async.wav")

    await client.stream_speech_to_file(
        "stream-async.wav",
        text="This file was written from the async streaming client.",
        voice_id=voice_id,
    )

    await client.close()


asyncio.run(main())

Available Methods

• client.health()
• client.list_voices()
• client.generate_speech(...)
• client.generate_speech_to_file(...)
• client.stream_speech(...)
• client.stream_speech_to_file(...)
• AsyncOuteAI.health()
• AsyncOuteAI.list_voices()
• AsyncOuteAI.generate_speech(...)
• AsyncOuteAI.generate_speech_to_file(...)
• AsyncOuteAI.stream_speech(...)
• AsyncOuteAI.stream_speech_to_file(...)

Error Handling

API errors raise OuteAIAPIError.

from outeai import OuteAI, OuteAIAPIError

client = OuteAI(api_key="YOUR_OUTEAI_API_KEY")

try:
    client.list_voices()
except OuteAIAPIError as exc:
    print(exc.status_code)
    print(exc.code)
    print(exc.message)

Notes

• TTS responses are returned as WAV audio.
• The default TTS model is outetts_1_pro.
• voice_id is required for generation and must belong to the authenticated user.