Transcribe

Transcribe

Turn audio into text via the OpenAI-compatible endpoint. Three models: gpt-4o-transcribe (most accurate), gpt-4o-mini-transcribe (cheaper), and whisper (billed per minute).

POST https://api.nexotao.com/v1/audio/transcriptions   (multipart/form-data)

Example

curl https://api.nexotao.com/v1/audio/transcriptions \
  -H "Authorization: Bearer sk-nexo-..." \
  -F file=@meeting.mp3 \
  -F model=gpt-4o-transcribe \
  -F language=id
from openai import OpenAI
 
client = OpenAI(base_url="https://api.nexotao.com/v1", api_key="sk-nexo-...")
 
with open("meeting.mp3", "rb") as f:
    t = client.audio.transcriptions.create(
        model="gpt-4o-transcribe",
        file=f,
        language="id",
    )
print(t.text)
import OpenAI from "openai"
import fs from "fs"
 
const client = new OpenAI({ baseURL: "https://api.nexotao.com/v1", apiKey: "sk-nexo-..." })
 
const t = await client.audio.transcriptions.create({
  model: "gpt-4o-transcribe",
  file: fs.createReadStream("meeting.mp3"),
  language: "id",
})
console.log(t.text)

Parameters & file limits

ParameterRequiredDescription
fileyesaudio file (mp3, wav, m4a, webm, mp4, mpeg, mpga, ogg, flac)
modelyesgpt-4o-transcribe, gpt-4o-mini-transcribe, or whisper
languagenoISO code, e.g. id for Indonesian — speeds up and sharpens results

Maximum file size: 25 MB.

Billing

  • gpt-4o-transcribe / mini: per audio token + output text tokens.
  • whisper: per minute of audio duration.

Exact rates are in Models & Pricing. The balance is deducted after the transcript is complete.

Image GenerationClaude Code