Audio

Text to speech

POST/v1/audio/speech

input is required, model defaults to tts-1, and voice picks the voice. The response is binary audio/mpeg, not JSON.

Request

curl https://api-dev.satulink.ai/v1/audio/speech \
  -H "Authorization: Bearer $SATULINK_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "model": "tts-1", "input": "Selamat pagi!", "voice": "alloy" }' \
  -o speech.mp3

import fs from "node:fs/promises";

// client dibuat seperti di bagian SDK
const speech = await client.audio.speech.create({
  model: "tts-1",
  voice: "alloy",
  input: "Selamat pagi!",
});

await fs.writeFile("speech.mp3", Buffer.from(await speech.arrayBuffer()));

# client dibuat seperti di bagian SDK
with client.audio.speech.with_streaming_response.create(
    model="tts-1",
    voice="alloy",
    input="Selamat pagi!",
) as response:
    response.stream_to_file("speech.mp3")

Speech to text

POST/v1/audio/transcriptions

Send multipart/form-data containing file (required) and model (defaults to whisper-1).

Request

curl https://api-dev.satulink.ai/v1/audio/transcriptions \
  -H "Authorization: Bearer $SATULINK_API_KEY" \
  -F [email protected] \
  -F model=whisper-1

import fs from "node:fs";

// client dibuat seperti di bagian SDK
const transcription = await client.audio.transcriptions.create({
  file: fs.createReadStream("audio.mp3"),
  model: "whisper-1",
});

console.log(transcription.text);

# client dibuat seperti di bagian SDK
with open("audio.mp3", "rb") as audio:
    transcription = client.audio.transcriptions.create(
        file=audio,
        model="whisper-1",
    )

print(transcription.text)

200 OK

json

{
  "text": "Selamat pagi!",
  "duration": 1.2,
  "language": "id",
  "usage": { "cost_idr": 5 }
}

No audio model is active yet. Requests fail until an administrator adds a route.

Still haven't found the answer? Try it directly in Playground or view the list of active models.