Developers

Your first API call takes five minutes.

Fourier TTS speaks, Fourier STT listens and Mira LLM replies, all on one key.

No sign-up to look around. Your sandbox starts with ₹200 of credit.

First audio, one request · POST /v1/audio/speech

curl https://sandbox.voice.miraiminds.co/v1/audio/speech \
  -H "Authorization: Bearer $MIRA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "mira-tts",
    "voice": "ashu",
    "response_format": "wav",
    "input": "नमस्ते, मैं आशु बोल रहा हूं। आपका ऑर्डर तैयार है।"
  }' \
  --output speech.wav

from openai import OpenAI

client = OpenAI(
    api_key="sk_live_...",
    base_url="https://sandbox.voice.miraiminds.co/v1",
)

with client.audio.speech.with_streaming_response.create(
    model="mira-tts",
    voice="ashu",
    response_format="pcm",
    input="चलिए numbers से start करते हैं।",
) as response:
    for chunk in response.iter_bytes():
        ...  # raw s16le mono @ 48 kHz

OpenAI’s speech protocol: change the base URL and the model name, keep the rest of your code.

The APIs

Three APIs give you the voice, the ears and the replies.

Each is priced on its own, by the character, the hour or the token.

Text to speech

Fourier TTS

₹1.80per 1,000 characters²

Monthly: No commitment
At once: 5 concurrent requests

Monthly: ₹50,000, credited to your wallet
At once: 25 concurrent requests

Shared endpoint · No contractual SLA. Top-ups use the same rate.
Enterprise: Custom volume pricing, concurrency, dedicated endpoint and SLA.

Read the TTS quickstart

Speech to text

Fourier STT

₹18an hour of recorded audio²

Offline transcription (POST /v2/transcriptions) is charged per second of audio, once, when a job completes. Degraded and failed jobs are free.
Speaker-labelled transcript with a speaker id per voice · Recordings up to 2 hours · Shared endpoint · No contractual SLA.

Read the transcription guide

Language model

Mira LLM

Gemma 4 26B A4B

₹15input · $0.15²

₹9.30cached input · $0.093

₹55output · $0.55, per 10,00,000 tokens

Monthly: No commitment
Rate limit: 1,000 requests a minute

Monthly: ₹50,000, credited to your wallet
Rate limit: 5,000 requests a minute

Google’s open Gemma 4 26B A4B, served on our own GPUs in India.
Rate limits apply per workspace across all API keys. Shared endpoint · No contractual SLA.
Enterprise: Custom volume pricing, rate limits, dedicated endpoint and SLA.

Read the LLM pricing

Quickstart

The quickstart is four requests and one real phone call.

It takes five minutes: a key, a number you may call, a public HTTPS URL.

Create a key
Create an agent
Place a call
Receive the webhook

Then put the three together on a call POST /v2/calls

from mirai import Mirai

# Uses MIRAI_API_KEY
mirai = Mirai()
call = mirai.calls.create(
    agent_id="YOUR_AGENT_ID",
    to="+91XXXXXXXXXX",
)

curl https://api.voice.miraiminds.co/v2/calls \
  -H "Authorization: Bearer $MIRAI_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: YOUR_UNIQUE_CALL_ID" \
  -d '{
    "agent_id": "YOUR_AGENT_ID",
    "to": "+91XXXXXXXXXX"
  }'

Use an agent you’ve configured in your workspace and a number you are allowed to call. For webhooks, ngrok is fine while you build.

Open the sandbox Open the quickstart

Questions

Questions developers ask before they build.

Which languages do the APIs support?

Hindi, English, Gujarati, Tamil and Telugu are live in Fourier TTS and Fourier STT. Fourier TTS also code-switches between Devanagari Hindi and Roman English in one sentence.

How is text to speech billed?

₹1.80 per 1,000 characters from a prepaid wallet. Pay as you go has 5 concurrent requests; Business has 25, with a ₹50,000 monthly commitment. All prices exclude GST.

Do I need to talk to sales?

No. Open the sandbox, add an email when you create a key, and start with ₹200 of credit. Talk to us when you need a dedicated endpoint or an SLA.

Does it work with the OpenAI SDK?

Fourier TTS speaks OpenAI’s POST /v1/audio/speech protocol, so the OpenAI SDKs work with a new base URL and model name. The same key works for the rest of the v2 API.

Still deciding? Ask us.

Fifteen minutes with the team that builds it. Bring your script, or your hardest call.

Book 15 minutes