Audio models

Text to speech, speech to text and realtime voice through one API: the requests, the voices, the formats, and the prices.

Available audio models

characters of input text

$15.00 / 1M characters

audio file transcription (batch)

$0.10 / hour

audio in and out, per minute of session

$0.08 / minute

audio in and out, per minute of session

$0.08 / minute

text input

$0.50 / 1M tokens

generated audio output

$10.00 / 1M tokens

text input

$1.00 / 1M tokens

generated audio output

$20.00 / 1M tokens

text input

$1.00 / 1M tokens

generated audio output

$20.00 / 1M tokens

text input

$1.00 / 1M tokens

generated audio output

$20.00 / 1M tokens

text input

$0.50 / 1M tokens

generated audio output

$9.00 / 1M tokens

text input

$0.50 / 1M tokens

generated audio output

$6.00 / 1M tokens

audio input

$2.00 / 1M tokens

transcript output

$12.00 / 1M tokens

audio input

$0.30 / 1M tokens

cached audio input

$0.03 / 1M tokens

transcript output

$2.50 / 1M tokens

audio input

$1.00 / 1M tokens

cached audio input

$0.10 / 1M tokens

transcript output

$3.00 / 1M tokens

Text to speech

curl -sS https://api.oneprovider.dev/v1/audio/speech \
-H "Authorization: Bearer sk-..." \
-H "Content-Type: application/json" \
-d '{
"model": "grok-tts",
"input": "Hello from OneProvider.",
"voice": "eve",
"response_format": "mp3"
}' \
--output hello.mp3

from openai import OpenAI
client = OpenAI(base_url="https://api.oneprovider.dev/v1", api_key="sk-...")
with client.audio.speech.with_streaming_response.create(
model="gemini-3.8-flash-tts",
voice="Kore",
input="Hello from OneProvider.",
response_format="wav",
) as response:
response.stream_to_file("hello.wav")

The reply is the audio file in the chosen format. For long text, send one request per paragraph and join the files.

Grok voices

Gemini voices

Speech to text

curl -sS https://api.oneprovider.dev/v1/audio/transcriptions \
-H "Authorization: Bearer sk-..." \
-F model=grok-stt \
-F file=@meeting.mp3
# {"text":"..."}

from openai import OpenAI
client = OpenAI(base_url="https://api.oneprovider.dev/v1", api_key="sk-...")
with open("meeting.mp3", "rb") as audio:
result = client.audio.transcriptions.create(model="gemini-3.5-transcribe", file=audio)
print(result.text)

The request is multipart/form-data with the fields file (the audio file) and model. The reply is JSON with a text field.

Realtime voice

Open a WebSocket to wss://api.oneprovider.dev/v1/realtime?model=grok-voice-3.0 with the same key in the Authorization header. The events follow the OpenAI Realtime API.

import asyncio, base64, json
import websockets
async def main():
url = "wss://api.oneprovider.dev/v1/realtime?model=grok-voice-3.0"
async with websockets.connect(url, additional_headers={"Authorization": "Bearer sk-..."}) as ws:
await ws.send(json.dumps({"type": "session.update", "session": {"instructions": "Answer briefly."}}))
await ws.send(json.dumps({"type": "conversation.item.create", "item": {
"type": "message", "role": "user",
"content": [{"type": "input_text", "text": "Say hello in one sentence."}],
}}))
await ws.send(json.dumps({"type": "response.create"}))
audio = bytearray()
async for message in ws:
event = json.loads(message)
if event["type"] == "response.output_audio.delta":
audio += base64.b64decode(event["delta"])
elif event["type"] == "response.output_audio_transcript.delta":
print(event["delta"], end="")
elif event["type"] == "response.done":
break
asyncio.run(main())

Pricing

• Grok TTS is billed per character of input text.
• Grok STT is billed per hour of audio.
• Grok Voice is billed per minute of conversation, plus a small fee for every text message you send.
• Gemini is billed in tokens: input text or audio, and output audio or text; one second of audio is 25 tokens.

Example: grok-tts, 1,000 characters = $15.00 / 1M × 1 000 = $0.015.

The exact price of every model is in the table above and on the pricing page .

Errors

Quick answers

Which base URL do I use?

Claude Code and Anthropic SDKs use https://api.oneprovider.dev. OpenAI SDKs and OpenAI-compatible clients use https://api.oneprovider.dev/v1. One key works behind both; the model list is on the models page ; rates are on the pricing page .

How do I authenticate?

Send the raw sk-… key in x-api-key or Authorization: Bearer: one of them, never both. Keep the key in an environment variable, not in a file that gets committed. More on authentication .

Where do I get a key?

Open your cabinet in the Telegram bot @oneprovider_robot or at app.oneprovider.dev, top up the balance and press “Issue a key”. Each key can carry its own spend limit. Cabinet & API keys .

What do 401, 404 and 429 mean?

401: the key is missing, mistyped or sent in the wrong header; paste the whole key without quotes or spaces. 404: usually a wrong base URL, for example a missing or doubled /v1. 429 with “Insufficient balance”: the cabinet balance or the key’s spend limit is used up, so top up instead of retrying; any other 429 is a rate limit, so wait for Retry-After. Details: error codes and troubleshooting .

Why do streams cut off or reset?

VPN and VLESS proxies often buffer SSE or break HTTP/2 negotiation. OneProvider has no regional restrictions and does not buffer streams, so turn the proxy off and retry.

Where do I ask if this is not enough?

Message @Oneproviderhelpbot on Telegram or open “Support” in the web cabinet, with the HTTP status code and the exact error message. How support works .