AI Text-to-Speech API - Runcrate

Convert text to natural-sounding speech with a single API call. Choose from lightweight models for real-time applications or larger models for studio-quality output.

Available TTS models

Model	Parameters	Strengths
`hexgrad/Kokoro-82M`	82M	Ultra-fast, low-latency, good for real-time apps
`canopylabs/orpheus-3b-0.1-ft`	3B	Expressive speech with emotion control
`Qwen/Qwen3-TTS`	—	Multilingual, high-quality prosody

Generate speech

curl https://api.runcrate.ai/v1/audio/speech \
  -H "Authorization: Bearer rc_live_YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "hexgrad/Kokoro-82M",
    "input": "Welcome to Runcrate. Your GPU instance is ready.",
    "voice": "af_heart"
  }' \
  --output welcome.mp3

from openai import OpenAI
from pathlib import Path

client = OpenAI(
    base_url="https://api.runcrate.ai/v1",
    api_key="rc_live_YOUR_API_KEY",
)

response = client.audio.speech.create(
    model="hexgrad/Kokoro-82M",
    input="Welcome to Runcrate. Your GPU instance is ready.",
    voice="af_heart",
)

Path("welcome.mp3").write_bytes(response.content)
print("Saved welcome.mp3")

from runcrate import Runcrate

client = Runcrate(api_key="rc_live_YOUR_API_KEY")

audio = client.models.generate_speech(
    model="hexgrad/Kokoro-82M",
    input="Welcome to Runcrate. Your GPU instance is ready.",
    voice="af_heart",
)

audio.save("welcome.mp3")

Expressive speech with Orpheus

Orpheus 3B supports emotion tags for more expressive output:

from openai import OpenAI
from pathlib import Path

client = OpenAI(
    base_url="https://api.runcrate.ai/v1",
    api_key="rc_live_YOUR_API_KEY",
)

response = client.audio.speech.create(
    model="canopylabs/orpheus-3b-0.1-ft",
    input="I just got the results back, and — honestly — I could not be happier.",
    voice="tara",
)

Path("expressive.mp3").write_bytes(response.content)

Multilingual speech with Qwen TTS

Qwen/Qwen3-TTS handles multiple languages automatically — pass text in any supported language and the model detects it:

from openai import OpenAI
from pathlib import Path

client = OpenAI(
    base_url="https://api.runcrate.ai/v1",
    api_key="rc_live_YOUR_API_KEY",
)

response = client.audio.speech.create(
    model="Qwen/Qwen3-TTS",
    input="Bienvenido a nuestra plataforma. Su cuenta está lista.",
)
Path("spanish.mp3").write_bytes(response.content)

Next steps

Models API — Text-to-Speech — full parameter reference and supported voices.
Batch transcription — go the other direction: speech to text at scale.
Model catalog — browse all available models.