Text-to-Speech | OpenDunes Docs

Text-to-Speech (TTS) generates spoken audio from text. OpenDunes supports Arabic including Modern Standard Arabic and Darija-inflected voices, making it suitable for voice assistants, IVR systems, and accessibility features targeting Algerian users.

Generate speech#

Python

import os
from openai import OpenAI
from pathlib import Path

client = OpenAI(
    base_url="https://opendunes.com/api/v1",
    api_key=os.environ["OPENDUNES_API_KEY"],
)

resp = client.audio.speech.create(
    model="x-ai/grok-voice-tts-1.0",
    voice="nova",
    input="مرحباً! أهلاً بك في OpenDunes، بوابتك إلى أكثر من مئتي نموذج ذكاء اصطناعي.",
)

Path("output.mp3").write_bytes(resp.content)

Stream to a file#

For longer texts, stream directly to disk to avoid buffering the full response in memory:

Python

with client.audio.speech.with_streaming_response.create(
    model="x-ai/grok-voice-tts-1.0",
    voice="alloy",
    input="OpenDunes routes your requests to the best available model and bills you in Algerian Dinars.",
) as response:
    response.stream_to_file("output.mp3")

Parameters#

Parameter	Description
`model`	A TTS model slug — filter the catalog with `?modality=speech`.
`voice`	Voice preset name — available voices vary by model.
`input`	Text to synthesize (up to the model's character limit).
`response_format`	`mp3` (default), `opus`, `aac`, `flac`, `wav`, `pcm`.
`speed`	0.25–4.0. Default `1.0`.

TTS is billed in DA per character of input text. Rates appear in the model catalog.

Generate speech#

Python

import os
from openai import OpenAI
from pathlib import Path

client = OpenAI(
    base_url="https://opendunes.com/api/v1",
    api_key=os.environ["OPENDUNES_API_KEY"],
)

resp = client.audio.speech.create(
    model="x-ai/grok-voice-tts-1.0",
    voice="nova",
    input="مرحباً! أهلاً بك في OpenDunes، بوابتك إلى أكثر من مئتي نموذج ذكاء اصطناعي.",
)

Path("output.mp3").write_bytes(resp.content)

Stream to a file#

For longer texts, stream directly to disk to avoid buffering the full response in memory:

Python

with client.audio.speech.with_streaming_response.create(
    model="x-ai/grok-voice-tts-1.0",
    voice="alloy",
    input="OpenDunes routes your requests to the best available model and bills you in Algerian Dinars.",
) as response:
    response.stream_to_file("output.mp3")

Parameters#

Parameter	Description
`model`	A TTS model slug — filter the catalog with `?modality=speech`.
`voice`	Voice preset name — available voices vary by model.
`input`	Text to synthesize (up to the model's character limit).
`response_format`	`mp3` (default), `opus`, `aac`, `flac`, `wav`, `pcm`.
`speed`	0.25–4.0. Default `1.0`.

TTS is billed in DA per character of input text. Rates appear in the model catalog.