Hindi TTS API Documentation for Developers

By NavVoice editorial team · Updated 2026-11-18

This documentation helps automotive engineering teams generate high-quality Hindi text-to-speech audio for in-car navigation and infotainment systems. You will find concrete code examples, parameter details, and workflow guides to integrate NavVoice into your multilingual routing logic.

We provide clear instructions for using curl or the official OpenAI SDK with our API base. Learn how to handle Hindi text, select voices, manage rate limits, and troubleshoot common issues like character limits or audio generation delays without unnecessary complexity.

Hindi TTS API Documentation for Developers

Frequently asked questions

How do I get an API key?

Sign up on the /get-api-key/ page using Google or email and password. Accounts are per site. The page displays your key, balance, trial, characters left, usage and a ready curl example.

Can I use the OpenAI SDK?

Yes. The request shape matches the OpenAI speech endpoint. Use the official OpenAI SDKs by changing the base_url to https://api.play3666.com/v1 and providing your NavVoice key.

What is the base_url for the API?

The API base URL is https://api.play3666.com/v1. Use this for all requests, including POST /v1/audio/speech, GET /v1/voices, and GET /v1/models.

Which audio formats are supported?

Supported formats are mp3, wav, opus, flac, and pcm. Specify the desired format in the response_format field of your request body.

How do I set the language to Hindi?

Include the optional language field with the ISO code "hi" in your JSON body. This ensures correct pronunciation for Hindi text, which uses the Devanagari script.

What is the rate limit per minute?

The limit is 60 requests per minute per key. New accounts start with 10 requests per minute until the first top-up is processed.

Does the tts-1-hd model cost more?

No. Both tts-1 and tts-1-hd use the same engine and price: $10 per 1 million characters. Choose based on quality preference, not cost.

How do I handle rate limit errors?

Implement retry logic in your application. Since requests are processed sequentially, queue your synthesis tasks and retry after the rate limit window resets.

Is the API key per account?

Yes, there is one active key per account. You can issue a new key at any time, which immediately invalidates the old one.

Can I regenerate my API key?

Yes. You can issue a new key from your dashboard at any time. The previous key stops working immediately upon regeneration.

What happens if the audio generation fails?

Failed requests are free. The charge is returned in full if synthesis fails, so you are not billed for unsuccessful audio generation attempts.

How do I send a JSON request?

Send a POST request to /v1/audio/speech with a JSON body containing model, input, voice, response_format, and optional speed or language fields.

Can I use the API for long texts?

Yes, but split text into parts of up to 4096 characters per request. Generate audio for each part and join the files manually.

Do I need to install any extra libraries?

No. Use the official OpenAI SDK or plain HTTP. No custom SDK, mobile SDK, or WordPress plugin is required to integrate with the API.

First request

curl

curl https://api.play3666.com/v1/audio/speech \
  -H "Authorization: Bearer $TTS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"tts-1","voice":"aria","input":"Hello from the API.","response_format":"mp3"}' \
  --output hello.mp3

Python (openai)

from openai import OpenAI

client = OpenAI(base_url="https://api.play3666.com/v1", api_key="tts-...")

with client.audio.speech.with_streaming_response.create(
    model="tts-1", voice="nova", input="Hello from Python.", response_format="mp3"
) as response:
    response.stream_to_file("hello.mp3")

Node.js (openai)

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

const client = new OpenAI({ baseURL: "https://api.play3666.com/v1", apiKey: process.env.TTS_API_KEY });
const res = await client.audio.speech.create({ model: "tts-1", voice: "milo", input: "Hello from Node.js." });
fs.writeFileSync("hello.mp3", Buffer.from(await res.arrayBuffer()));
voice
ariafemale
novafemale
lunafemale
sagefemale
felixmale
milomale
atlasmale
orionmale
language
arالعربية
zh中文
daDansk
nlNederlands
enEnglish
fiSuomi
frFrançais
deDeutsch
elΕλληνικά
heעברית
hiहिन्दी
itItaliano
ja日本語
ko한국어
msBahasa Melayu
noNorsk
plPolski
ptPortuguês
ruРусский
esEspañol
swKiswahili
svSvenska
trTürkçe

Getting Your API Key

Sound wave illustration for: Getting Your API Key

NavVoice requires an API key to authenticate your requests. Sign up on https://play3666.com/get-api-key/ using Google or an email account. This creates a unique account specific to this site.

Sign Up on the Site

Visit the sign-up page to create your account. You can authenticate via Google or use a standard email and password combination. Accounts are tied to this specific storefront.

Retrieve Your Key

After signing in, the dashboard displays your active API key. You can also issue a new key at any time, which immediately invalidates the previous one.

  • One active key per account
  • New keys invalidate the old one instantly

Authorization Header

Include your key in every request using the standard Bearer token format. This ensures your usage is tracked against your balance.

  • Format: Authorization: Bearer tts-...

First Request with Curl

Sound wave illustration for: First Request with Curl

To generate speech, send a POST request to https://api.play3666.com/v1/audio/speech with your API key in the Authorization header. The body must include the model, input text, voice, and response format.

Basic JSON Body

Include the required fields: model (tts-1 or tts-1-hd), input (your text), voice (one of the eight presets), and response_format (mp3, wav, opus, flac, or pcm).

  • Model: tts-1 or tts-1-hd
  • Voice: e.g., aria, nova, luna
  • Format: mp3, wav, opus, flac, pcm

Sending Hindi Text

Pass your Hindi text in the input field. You can optionally set the language parameter to 'hi' to ensure correct pronunciation, though the system often guesses Latin-script languages as English.

Receiving Audio

The API returns the audio file directly in the response body. For Hindi navigation cues, ensure your client handles binary data correctly to save the file locally or play it on your in-car system.

First Request with Curl

Using the OpenAI SDK

Sound wave illustration for: Using the OpenAI SDK

The NavVoice API mirrors the OpenAI speech endpoint structure, allowing you to reuse existing client libraries without rewriting core logic. By simply adjusting the base URL, your team can integrate high-quality Hindi TTS into automotive navigation systems using familiar code patterns.

This approach minimizes development time and reduces boilerplate code, enabling large automotive teams to deploy voice features quickly across infotainment platforms.

Change base_url

Point your SDK client to https://api.play3666.com/v1 instead of the default OpenAI endpoint. This redirects all requests to the NavVoice infrastructure while maintaining standard authentication and payload formats.

  • Base URL: https://api.play3666.com/v1

Use Official SDKs

Leverage the official OpenAI SDKs (Python, Node.js, etc.) directly. No proprietary client library is required, ensuring compatibility with your existing engineering workflows and CI/CD pipelines.

Same Request Shape

The JSON body structure remains identical to the OpenAI speech endpoint. Pass your model, input, voice, and response_format fields exactly as you would for OpenAI, but they will be processed by NavVoice engines.

Supported Audio Formats

Sound wave illustration for: Supported Audio Formats

NavVoice supports five standard audio formats to suit different playback environments and storage constraints. You can specify your preferred output type using the response_format parameter in your API request. The service returns the synthesized audio file directly in the chosen format.

MP3 and Opus

MP3 is the universal standard for web and mobile playback, offering small file sizes with good quality. Opus provides superior compression efficiency, making it ideal for bandwidth-constrained in-car infotainment systems or long navigation instructions where storage space is limited.

  • MP3: Broad compatibility, moderate file size.
  • Opus: High efficiency, ideal for low-bandwidth transmission.

WAV and FLAC

WAV delivers uncompressed audio, ensuring zero quality loss for critical safety announcements. FLAC offers lossless compression, reducing file size by approximately 50–60% compared to WAV while preserving exact audio fidelity, perfect for offline route databases.

  • WAV: Uncompressed, simple decoding.
  • FLAC: Lossless compression, smaller storage footprint.

PCM Raw Audio

PCM provides raw, uncompressed 16-bit signed integer samples. This format is optimal for direct hardware integration or when your application requires precise control over audio processing pipelines, eliminating the need for additional decoding steps.

  • PCM: Raw 16-bit signed integers.
  • Use case: Direct hardware or DSP integration.

Voice and Language Parameters

Sound wave illustration for: Voice and Language Parameters

NavVoice provides eight distinct voice presets optimized for automotive clarity, allowing your team to select tones that enhance driver experience across diverse in-car environments. Each voice is engineered for crisp audio delivery, ensuring navigation instructions remain intelligible even in noisy cabin conditions.

While every voice supports all 23 languages, explicitly setting the language parameter prevents misidentification for non-English Latin-script text, ensuring accurate pronunciation for Hindi, Arabic, or other supported languages.

Selecting a Voice

Choose from four female voices (Aria, Nova, Luna, Sage) and four male voices (Felix, Milo, Atlas, Orion). Select Aria for warm narration or Atlas for neutral news-style updates. All voices are derived from licensed public-domain recordings, ensuring high-quality, consistent synthetic speech.

  • Aria: Warm, clear narrator
  • Nova: Bright, upbeat
  • Luna: Soft, storytelling
  • Sage: Calm, instructional
  • Felix: Energetic, promo
  • Milo: Friendly, conversational
  • Atlas: Neutral, news
  • Orion: Deep, documentary

Setting the Language

The API supports 23 languages, including Hindi, Arabic, English, and German. You can pass an optional language ISO code to override automatic detection. This is critical for mixed-script texts or when using Latin script for languages like Hindi or Turkish to ensure correct phonetic rendering.

  • Supports 23 languages
  • Optional language parameter
  • Overrides script-based guessing
  • Ensures correct pronunciation
  • Not supported: Vietnamese, Thai, Indonesian, Ukrainian, Romanian, Czech, Bengali
  • Not supported: Ukrainian, Romanian

Speed Control

Adjust the speed parameter between 0.25 and 4.0 to fit navigation pacing. Slower speeds improve clarity for complex directions, while higher speeds suit quick alerts. Note that requests are processed sequentially, so under load a request may wait in a queue.

Voice and Language Parameters

Rate Limits and Errors

Sound wave illustration for: Rate Limits and Errors

The API enforces a strict rate limit of 60 requests per minute per key, allowing up to 10 requests per minute before your first top-up.

60 Requests Per Minute

Each API key is limited to 60 requests per minute. Before your initial top-up, this limit is reduced to 10 requests per minute. If your in-car navigation system processes long texts, remember to split them into chunks of 4096 characters to stay within these bounds.

Handling 429 Errors

When you exceed the limit, the API returns a 429 Too Many Requests error. Implement exponential backoff in your integration to retry after a short delay. This is common when generating audio for complex, multi-language routes during peak usage hours.

  • Return code 429 indicates rate limit exceeded.
  • Retry logic should include exponential backoff.
  • Monitor your usage dashboard for real-time limits.

Failed Request Refunds

If synthesis fails due to a server error, the charge is returned in full. You are only billed for successfully generated audio files. This ensures your team pays exactly for usable navigation audio without hidden costs for failed attempts.

Model Options

Sound wave illustration for: Model Options

NavVoice offers two model variants for Hindi TTS generation, both powered by the same high-quality engine. The choice depends on your preference for processing speed versus audio fidelity, though the underlying voice characteristics remain consistent across both options.

tts-1 Model

The standard "tts-1" model provides efficient synthesis with a focus on speed. It is ideal for high-volume navigation prompts where real-time generation is critical. Text is processed quickly, making it suitable for dynamic route updates in infotainment systems.

tts-1-hd Model

The "tts-1-hd" model delivers higher fidelity audio with enhanced naturalness. Use this for premium brand experiences or static voiceovers where audio quality is prioritized over raw generation speed. The output is richer but may take slightly longer to process.

Same Price for Both

Both models are priced identically at $10 per 1 million characters. You can switch between "tts-1" and "tts-1-hd" in your API calls without any cost difference. This allows teams to optimize for quality or speed based on specific use cases without budget impact.

Model Options

Troubleshooting

Sound wave illustration for: Troubleshooting

When integrating NavVoice for automotive navigation, developers often encounter specific HTTP errors or latency issues. Understanding these common pitfalls ensures smooth audio generation for in-car systems without unexpected interruptions.

Common HTTP Errors

A 401 Unauthorized error indicates an invalid or expired API key. Ensure your Authorization header uses the correct Bearer tts-... format. A 429 Too Many Requests error means you have exceeded the 60 requests per minute limit. Reduce your request frequency or implement exponential backoff in your automotive app logic.

  • 401: Check key format
  • 429: Wait before retrying

Audio Generation Delays

Requests are processed sequentially. Under load, a request may wait in a queue, causing slight delays. For long navigation instructions, this is usually negligible, but short phrases might take a few seconds. If your system requires real-time responses, consider pre-generating common phrases.

Character Limit Issues

Each request supports up to 4096 characters. If your navigation text exceeds this, split the text into smaller parts, generate audio for each, and join the files. Ensure spaces and punctuation count toward the limit. Failed requests do not incur charges, so retrying is safe if the error is transient.

Listen to the 8 voices

Each sample below was generated by this API from the same sentence. Our navigation system ensures precise routing for every vehicle in the fleet.

Avatar of the voice aria
ariafemale
Avatar of the voice nova
novafemale
Avatar of the voice luna
lunafemale
Avatar of the voice sage
sagefemale
Avatar of the voice felix
felixmale
Avatar of the voice milo
milomale
Avatar of the voice atlas
atlasmale
Avatar of the voice orion
orionmale

23 languages

Any voice speaks any of these languages.

Sources and further reading

  1. Wikipedia
  2. FFmpeg documentation
  3. Creative Commons Attribution 4.0
  4. MDN: Web Speech API
  5. curl documentation
  6. TRON developer hub: TRC-20
  7. LibriTTS-R paper (arXiv 2305.18802)
  8. ISO 639 language codes
  9. LibriTTS-R corpus (OpenSLR 141)
  10. openai-node on GitHub
  11. LibriVox public-domain audiobooks
  12. W3C WAI: making audio and video accessible

More on this site