/api/v1/audio/transcriptions

Speech to text. Upload an audio file as multipart/form-data and get the transcript back as JSON, plain text or subtitles. The request follows OpenAI's audio.transcriptions.create.

URL: https://railwail.com/api/v1/audio/transcriptions
Key scope: audio
Body: multipart, file ≤ 25 MB
Returns: json · text · srt · vtt

Form fields

filerequired

file

The recording: mp3, wav, m4a, webm, ogg, flac or mp4, at most 25 MB.

modelrequired

string

Model slug from the list below.

language

string

ISO-639-1 hint such as en or de. Improves accuracy when known.

prompt

stringmodel-dependent

Context to bias the decoder: names, jargon, the previous sentence.

response_format

string

json, verbose_json, text, srt or vtt. text, srt and vtt come back as plain text.Default json

temperature

numbermodel-dependent

0–1.

timestamp_granularities

string[]model-dependent

word and/or segment, with verbose_json.

Examples

Key in RAILWAIL_API_KEY with the audio scope; audio.mp3 is your file.

curl https://railwail.com/api/v1/audio/transcriptions \
  -H "Authorization: Bearer $RAILWAIL_API_KEY" \
  -F [email protected] \
  -F model=whisper-replicate \
  -F response_format=json

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["RAILWAIL_API_KEY"],
    base_url="https://railwail.com/api/v1",
)

with open("audio.mp3", "rb") as audio:
    transcript = client.audio.transcriptions.create(model="whisper-replicate", file=audio)
print(transcript.text)

// ESM: save as transcribe.mjs
import fs from "node:fs";
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.RAILWAIL_API_KEY,
  baseURL: "https://railwail.com/api/v1",
});

const transcript = await client.audio.transcriptions.create({
  model: "whisper-replicate",
  file: fs.createReadStream("audio.mp3"),
});
console.log(transcript.text);

Response

With json (default); verbose_json adds segments and, where the model returns them, language and duration.

JSON

{
  "text": "<the transcript>"
}

Subtitles: response_format=srt or vtt return the file content as text. The railwail npm SDK has no transcription method; use the OpenAI SDK or plain HTTP.

Models you can call here

2 Modelle laufen gerade über diesen Endpunkt · Preise wie berechnet, in USDAlle mit Filtern ansehen

Errors

Status	Code	What to do
400	`missing_file`	No file field in the form.
400	`empty_file`	The uploaded file has no content.
400	`invalid_body`	The body is not multipart/form-data.
413	`file_too_large`	Over 25 MB. Split the recording or compress it.
402	`insufficient_credits`	Top up on the billing page.
403	`insufficient_scope`	The key lacks the audio scope.
429	`trial_limit`	Trial rules (at most 2 credits per run until the first top-up).
503	`model_unavailable`	No verified price or the provider is not reachable; pick another model.

All codes: Error codes.

/api/v1/audio/transcriptions

Form fields#

Examples#

Response#

Models you can call here#

Errors#

Form fields

Examples

Response

Models you can call here

Errors