Audio (Speech & Transcription)

Provides text-to-speech (TTS) and speech-to-text (STT). Compatible with the OpenAI Audio API.

Prompt-based music generation via POST /v1/audio/generations is currently unavailable. See /docs/api/music for details.

Text-to-speech#

POST/v1/audio/speech

POST /v1/audio/speech converts text into spoken audio. Requests require an API key beginning with plm_.

Parameter	Type	Required	Description
model	string	Required	TTS model ID, e.g. `gemini-3.8-flash-tts`.
input	string	Required	Text to convert. Up to 8,000 bytes of UTF-8. ElevenLabs models are limited to 5,000 bytes, except `eleven_flash_v2` and `eleven_flash_v2_5`. Each Korean character takes 3 bytes.
voice	string	Optional	Voice name. Accepted values differ by model; omit it to use the model's default voice. `gemini-3.8-flash-tts` accepts prebuilt voice names such as `Kore`.
response_format	string	Optional	Audio format (e.g. `wav`, `mp3`). Supported formats differ by model; `gemini-3.8-flash-tts` responds with `wav`.

curl

curl https://apirouter.pleum.ai/v1/audio/speech \
  -H "Authorization: Bearer $PLEUM_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gemini-3.8-flash-tts",
    "input": "Hello from PleumRouter.",
    "voice": "Kore",
    "response_format": "wav"
  }' \
  --output out.wav

OpenAI SDK (Python)

import os
from openai import OpenAI

client = OpenAI(api_key=os.environ["PLEUM_API_KEY"], base_url="https://apirouter.pleum.ai/v1")

response = client.audio.speech.create(
    model="gemini-3.8-flash-tts",
    input="Hello from PleumRouter.",
    voice="Kore",
    response_format="wav",
)
response.write_to_file("out.wav")

OpenAI SDK (TypeScript)

import OpenAI from "openai";
import fs from "node:fs";

const client = new OpenAI({ apiKey: process.env.PLEUM_API_KEY, baseURL: "https://apirouter.pleum.ai/v1" });

const response = await client.audio.speech.create({
  model: "gemini-3.8-flash-tts",
  input: "Hello from PleumRouter.",
  voice: "Kore",
  response_format: "wav",
});

fs.writeFileSync("out.wav", Buffer.from(await response.arrayBuffer()));

The response body of this endpoint is raw audio bytes, not JSON. The Content-Type is set according to the format, e.g. audio/wav or audio/mpeg (mp3). Cost is returned in the response headers X-Cost-Krw and X-Cost-Usd, not in the body. The billing unit depends on the model (characters or tokens); check price_unit in the GET /v1/models response.

Transcription#

POST/v1/audio/transcriptions

POST /v1/audio/transcriptions transcribes an audio file into text. Requests require an API key beginning with plm_.

This endpoint is requested as multipart/form-data — you upload the file as a form field rather than sending a JSON body.

Parameter	Type	Required	Description
file	file	Required	Audio file to transcribe, uploaded as a form field. Accepts flac, mp4, mpeg, ogg, wav, and webm, up to 25MB. The Content-Type of the part must match the format, for example `audio/mpeg` for mp3. An unsupported type returns `415`, and a file over 25MB returns `413`.
model	string	Optional	Transcription model ID (form field). Default `whisper-1`.

curl

curl https://apirouter.pleum.ai/v1/audio/transcriptions \
  -H "Authorization: Bearer $PLEUM_API_KEY" \
  -F "file=@speech.mp3;type=audio/mpeg" \
  -F model=whisper-1

OpenAI SDK (Python)

import os
from openai import OpenAI

client = OpenAI(api_key=os.environ["PLEUM_API_KEY"], base_url="https://apirouter.pleum.ai/v1")

with open("speech.mp3", "rb") as f:
    response = client.audio.transcriptions.create(model="whisper-1", file=f)
print(response.text)

OpenAI SDK (TypeScript)

import OpenAI, { toFile } from "openai";
import fs from "node:fs";

const client = new OpenAI({ apiKey: process.env.PLEUM_API_KEY, baseURL: "https://apirouter.pleum.ai/v1" });

const response = await client.audio.transcriptions.create({
  model: "whisper-1",
  file: await toFile(fs.createReadStream("speech.mp3"), "speech.mp3", { type: "audio/mpeg" }),
});

console.log(response.text);

The response is JSON containing the transcribed text, the model used, and cost (KRW cost, FX rate, markup). Billing is per minute of audio.

200 OK

{
  "text": "Hello from PleumRouter.",
  "model": "whisper-1",
  "cost": {
    "usd": 0.006,
    "krw": 8.1,
    "fx_rate": 1342.48,
    "markup_rate": 0.0
  }
}