フリーAI 音声からテキストへ

import requests

response = requests.post(
    "https://api.stt.ai/v1/transcribe",
    headers={"Authorization": f"Bearer {API_KEY}"},
    files={"file": open("meeting.mp3", "rb")},
    data={
        "model": "large-v3-turbo",
        "language": "auto",
        "diarize": "true",
        "response_format": "json",
    },
)

result = response.json()
for seg in result["segments"]:
    print(f"{seg['speaker']}: {seg['text']}")

import fs from "fs";

const form = new FormData();
form.append("file", fs.createReadStream("meeting.mp3"));
form.append("model", "large-v3-turbo");
form.append("language", "auto");
form.append("diarize", "true");

const res = await fetch("https://api.stt.ai/v1/transcribe", {
  method: "POST",
  headers: { Authorization: `Bearer ${API_KEY}` },
  body: form,
});

const { segments } = await res.json();
segments.forEach(s =>
  console.log(`${s.speaker}: ${s.text}`)
);