Voice AI quickstart

This quickstart walks through creating a Voice AI API key, calling Speech-to-Text, and confirming the call lands in your usage dashboard.

1. Sign up and create a key.

   Open console.gocommotion.com, sign up, and create a Voice AI API key. Sandbox keys (eak_test_*) call the same endpoints with separate credit wallet for testing.

2. Make your first call.

   Save a short audio file as sample.wav (any of WAV, MP3, M4A, FLAC works) and call the Speech-to-Text endpoint:

   curl

   curl -X POST https://api.gocommotion.com/ai/speech-to-text \
     -H "X-API-Key: eak_live_your_key_here" \
     -F "file=@sample.wav" \
     -F "language=en"

   Python

   import requests
   
   with open("sample.wav", "rb") as audio:
       response = requests.post(
           "https://api.gocommotion.com/ai/speech-to-text",
           headers={"X-API-Key": "eak_live_your_key_here"},
           files={"file": audio},
           data={"language": "en"},
       )
   
   print(response.json()["text"])

   Node

   import { readFileSync } from "node:fs";
   
   const form = new FormData();
   form.append("file", new Blob([readFileSync("sample.wav")]), "sample.wav");
   form.append("language", "en");
   
   const response = await fetch(
     "https://api.gocommotion.com/ai/speech-to-text",
     {
       method: "POST",
       headers: { "X-API-Key": "eak_live_your_key_here" },
       body: form,
     },
   );
   
   const data = await response.json();
   console.log(data.text);

3. Check usage in the console.

   Open the Voice AI usage dashboard - the call should appear within a few seconds, with credits debited from your Voice AI wallet.

Tip

For long-running streams, use the Streaming guide instead. Speech-to-Text is for finished files.