Quickstart

curl https://api.eesi.ai/v1/audio/models \
  -H "Authorization: Bearer $EESI_API_KEY"

from openai import OpenAI

client = OpenAI(api_key="sk-eesi-...", base_url="https://api.eesi.ai/v1")

with client.audio.speech.with_streaming_response.create(
    model="nur-tts-v1",
    voice="Nur",
    input="The rain had not stopped for three days.",
) as response:
    response.stream_to_file("line.mp3")

with open("line.mp3", "rb") as audio:
    transcript = client.audio.transcriptions.create(
        model="nur-stt-v1",
        file=audio,
    )
print(transcript.text)

import asyncio, base64, json, websockets

URL = "wss://api.eesi.ai/v1/realtime?model=nur-realtime-v1"

async def main():
    async with websockets.connect(
        URL, additional_headers={"Authorization": "Bearer sk-eesi-..."}
    ) as ws:
        await ws.send(json.dumps({
            "type": "session.update",
            "session": {
                "type": "realtime",
                "instructions": "You are warm and concise. Keep replies short.",
                # The voice goes here and nowhere else.
                "audio": {"output": {"voice": "Nur"}},
            },
        }))

        # 16 kHz mono PCM16, base64, in ~20 ms chunks, paced in real time —
        # server-side turn detection reads arrival timing.
        await ws.send(json.dumps({
            "type": "input_audio_buffer.append",
            "audio": base64.b64encode(chunk).decode(),
        }))

        async for raw in ws:
            event = json.loads(raw)
            if event["type"] == "response.output_audio.delta":
                play(base64.b64decode(event["delta"]))
            elif event["type"] == "input_audio_buffer.speech_started":
                stop_playback()   # barge-in: drop what you have queued

asyncio.run(main())

curl "https://api.eesi.ai/v1/speech/sessions?limit=5" \
  -H "Authorization: Bearer $EESI_API_KEY"