Audio Transcription

You can transcribe audio using the Whisper Large-v3 model deployed on your Cube via an OpenAI-compatible API.

---

JavaScript

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://whisper-eukodyne.edgescaleai-cube.com/v1/"
});

const response = await client.audio.transcriptions.create({
  model: "openai/whisper-large-v3",
  file: "path/to/audio_sample.wav",
  language: "en"
});

console.log(response.text);

Python

from openai import OpenAI

client = OpenAI(
    api_key="none",
    base_url="https://whisper-eukodyne.edgescaleai-cube.com/v1/"
)

response = client.audio.transcriptions.create(
    model="openai/whisper-large-v3",
    file="audio_sample.wav",
    language="en"
)

print(response.text)

cURL

curl -X POST "https://whisper-eukodyne.edgescaleai-cube.com/v1/audio/transcriptions" \
  -F model="openai/whisper-large-v3" \
  -F file="@audio_sample.wav" \
  -F language="en"