Quickstart

runanything.ai serves two OpenAI-compatible endpoints — text-to-speech and transcription — at the base URL https://runanything.ai/v1. If you've used OpenAI's audio APIs, you already know this one: same request shapes, same responses, same SDKs.

1. Get an API key

We're in private beta and issue keys by hand. Email help@runanything.ai with a line about what you're building and your expected volume — we'll usually reply with a key within a day.

Request an API key

2. Generate speech

Returns mp3 by default (wav, aac, and streaming pcm are one parameter away — see Text to speech).

curl https://runanything.ai/v1/audio/speech \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "kokoro-82m",
    "input": "Hello! This is my first request.",
    "voice": "af_heart"
  }' \
  --output speech.mp3

from openai import OpenAI

client = OpenAI(
    base_url="https://runanything.ai/v1",
    api_key="YOUR_API_KEY",
)

speech = client.audio.speech.create(
    model="kokoro-82m",
    voice="af_heart",
    input="Hello! This is my first request.",
)
speech.write_to_file("speech.mp3")

import OpenAI from "openai";
import fs from "node:fs";

const client = new OpenAI({
  baseURL: "https://runanything.ai/v1",
  apiKey: "YOUR_API_KEY",
});

const speech = await client.audio.speech.create({
  model: "kokoro-82m",
  voice: "af_heart",
  input: "Hello! This is my first request.",
});
fs.writeFileSync("speech.mp3", Buffer.from(await speech.arrayBuffer()));

Voices: 28 built-ins like af_heart and bm_george (full list), and OpenAI names like nova work too.

3. Transcribe audio

Upload webm, mp4, ogg, wav, or mp3 (up to 4 MB during the beta) and get text back.

curl https://runanything.ai/v1/audio/transcriptions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F file=@recording.wav \
  -F model=distil-whisper-large-v3

result = client.audio.transcriptions.create(
    model="distil-whisper-large-v3",
    file=open("recording.wav", "rb"),
)
print(result.text)

const result = await client.audio.transcriptions.create({
  model: "distil-whisper-large-v3",
  file: fs.createReadStream("recording.wav"),
});
console.log(result.text);

Next steps

Text to speech — formats, speed, and streaming raw PCM for real-time playback.
Speech to text — response formats and language handling.
Errors & limits — what 4xx/5xx responses look like and current beta limits.