Transcriptions & Translations API

FastRouter provides OpenAI-compatible audio APIs for transcribing speech into text or translating audio into English, using the Whisper model.

Introduction

FastRouter provides OpenAI-compatible endpoints for converting audio into text — either as a direct transcription or as a translation into English. The API accepts common audio formats and supports multiple output types such as plain text, JSON, SRT, and VTT.

Before you begin, replace placeholders like <FASTROUTER_API_KEY> with your actual key (see Keys & Settings in your dashboard).

---

Models

FastRouter supports:

• openai/whisper-1

---

Transcriptions API

Use this endpoint to convert spoken audio into written text in the original language.

Endpoint

POST https://go.fastrouter.ai/api/v1/audio/transcriptions

Base URL

https://go.fastrouter.ai/api/v1

Supported Input File Formats

MP3, MP4, MPEG, MPGA, M4A, WAV, WEBM

Parameters

Parameter	Description
model	Must be "whisper-1"
file	Path to the audio file
language	(Optional) ISO-639-1 code (e.g., "en", "es")
response_format	(Optional) Output format (see below)
temperature	(Optional) Sampling temperature

---

Using the OpenAI SDK (Python)

from openai import OpenAI

client = OpenAI(
    api_key="<FASTROUTER_API_KEY>",
    base_url="https://go.fastrouter.ai/api/v1"
)

audio_file = open("audio.mp3", "rb")

response = client.audio.transcriptions.create(
    model="whisper-1",
    file=audio_file,
    language="en",
    response_format="verbose_json"
)

print(response)

---

Transcription Example (cURL)

curl --location 'https://go.fastrouter.ai/v1/audio/transcriptions' \
  --header 'Authorization: Bearer <FASTROUTER_API_KEY>' \
  --form 'model="whisper-1"' \
  --form 'language="en"' \
  --form 'file=@"/path/to/file.mp3"' \
  --form 'response_format="verbose_json"'

---

Output Formats

Set response_format to:

• json — Basic structured output

• text — Plain text

• srt — SubRip subtitle file

• vtt — WebVTT subtitle file

• verbose_json — Metadata + transcript

---

Translations API

Use this endpoint to translate audio into English, regardless of the source language. The API automatically transcribes and translates the audio.

Note: This endpoint always outputs English text.

Endpoint

POST /v1/audio/translations

---

Using the OpenAI SDK (Python)

from openai import OpenAI

client = OpenAI(
    api_key="<FASTROUTER_API_KEY>",
    base_url="https://go.fastrouter.ai/api/v1"
)

audio_file = open("speech_spanish.mp3", "rb")

response = client.audio.translations.create(
    model="whisper-1",
    file=audio_file,
    response_format="verbose_json",
    temperature=0.1
)

print(response)

---

Translation Example (cURL)

curl --location 'https://go.fastrouter.ai/v1/audio/translations' \
  --header 'Authorization: Bearer <FASTROUTER_API_KEY>' \
  --form 'model="whisper-1"' \
  --form 'file=@"/path/to/file.mp3"' \
  --form 'response_format="verbose_json"' \
  --form 'temperature="0.1"'