Sprachtranskription (POST /v1/audio/transcriptions)

Transkribiert Audio in Text im angegebenen Format mit vollständiger OpenAI-API-Wire-Kompatibilität.

---

HTTP-Anfrage

POST /v1/audio/transcriptions
Authorization: Bearer sk-oneasr-...
Content-Type: multipart/form-data

---

Anfrage-Body (Multipart Form)

Feld	Typ	Erforderlich	Beschreibung
file	binary	Ja	Die Audiodatei (mp3, mp4, wav, m4a, ogg, webm, flac usw.).
model	string	Ja	Ziel-Engine-/Modellname (z. B. faster-whisper, qwen, whisper-1).
language	string	Nein	ISO-639-1-Sprachcode (z. B. de, en, zh, ja, ko).
prompt	string	Nein	Optionaler Leit-Prompt zur Steuerung von Schreibweisen oder Fachbegriffen.
response_format	string	Nein	Ausgabeformat: json (Standard), text, srt, verbose_json, vtt.
temperature	number	Nein	Sampling-Temperatur (0.0 bis 1.0, Standard: 0.0).
timestamp_granularities	array	Nein	Zeitstempel-Granularität für verbose_json: ["word"], ["segment"].

---

Code-Beispiele

cURL

curl -X POST http://localhost:8000/v1/audio/transcriptions   -H "Authorization: Bearer sk-oneasr-v1-xxx"   -F file="@podcast.mp3"   -F model="faster-whisper"   -F response_format="verbose_json"

Python (Offizielles OpenAI SDK)

from openai import OpenAI

client = OpenAI(
    base_url="http://localhost:8000/v1",
    api_key="sk-oneasr-v1-xxx"
)

with open("speech.mp3", "rb") as f:
    result = client.audio.transcriptions.create(
        model="faster-whisper",
        file=f,
        response_format="verbose_json"
    )

print(result.text)

---

Antwort (verbose_json)

{
  "task": "transcribe",
  "language": "english",
  "duration": 5.24,
  "text": "Welcome to OneASR private voice gateway.",
  "segments": [
    {
      "id": 0,
      "seek": 0,
      "start": 0.0,
      "end": 5.2,
      "text": "Welcome to OneASR private voice gateway.",
      "tokens": [50364, 4843, 284, 1530, 50589],
      "temperature": 0.0,
      "avg_logprob": -0.12,
      "compression_ratio": 1.1,
      "no_speech_prob": 0.001
    }
  ]
}