Cloud Text-to-Speech API for Audio Generation

By BlogSpeak editorial team · Updated 2026-11-18

This documentation provides the technical reference for BlogSpeak’s cloud text-to-speech API, designed for freelancers and content creators who need reliable audio generation without complex infrastructure. You will find precise specifications for endpoints, parameters, and authentication to integrate synthetic voice assets directly into your WordPress blogs or podcast workflows.

The guide covers request shapes compatible with official OpenAI SDKs, rate limits, and error handling. It details the eight voice presets, supported output formats, and pricing structures. Use this reference to implement robust audio synthesis for your written content, ensuring seamless delivery of speech files generated from your text inputs.

Cloud Text-to-Speech API for Audio Generation

Frequently asked questions

How do I get an API key?

Visit /get-api-key/ to sign up using Google or email. Your unique key appears on the dashboard. One active key is issued per account, and you can generate a new one at any time, which immediately invalidates the previous key.

What is the base URL for the API?

The API base URL is https://api.playf44.com/v1. Use this endpoint for all requests, including POST /v1/audio/speech for generation and GET /v1/voices for metadata. This structure ensures compatibility with standard HTTP clients and SDKs.

Does it support the OpenAI SDK?

Yes. The request shape matches the OpenAI speech endpoint. You can use the official OpenAI SDK by changing the base_url to https://api.playf44.com/v1 and providing your BlogSpeak key. This allows seamless integration into existing workflows.

What formats can I receive audio in?

You can receive audio in mp3, wav, opus, flac, or pcm formats. Specify your preference using the response_format field in the JSON body. These formats cover most common use cases for blogs, podcasts, and applications.

What is the character limit per request?

Each request supports up to 4096 characters of input text. If your content exceeds this limit, split it into parts and join the resulting audio files. Spaces and punctuation count toward the total character count.

How fast is the API?

Generation takes roughly real time, processing about 25 characters per second for long texts. Short phrases may take a few seconds. Requests are processed sequentially, so high load may introduce queue delays.

What happens if I hit the rate limit?

The limit is 60 requests per minute per key (10 per minute before the first top-up). Exceeding this may result in delayed processing or errors. Monitor your usage via the dashboard to stay within limits.

Can I stream audio in real-time?

No. The API returns the complete audio file in the response. There is no streaming, WebSocket, or real-time partial audio support. You must wait for the full synthesis to complete before receiving the audio data.

Is SSML supported?

No. The API does not support SSML, phoneme dictionaries, or custom pronunciation lexicons. You can adjust speed using the speed parameter (0.25–4.0) and select from eight voice presets, but no fine-grained markup is available.

How do I handle long texts?

Split your text into chunks of up to 4096 characters. Send each chunk as a separate POST request and join the returned audio files. This ensures reliable synthesis and avoids exceeding the per-request character limit.

What is the audio response format?

The response contains the audio file in your requested format (mp3, wav, opus, flac, or pcm). Failed requests return no charge. Each file includes an inaudible watermark to identify it as synthetic speech.

Can I use my own voice model?

No. The API offers eight preset voices based on licensed recordings. Custom voice cloning, uploading your own voice, or voice design are not supported. You can choose from female and male presets for different tones.

Is there a WebSocket endpoint?

No. There is no WebSocket endpoint for real-time audio streaming. The API operates over standard HTTP POST requests. You receive the full audio file upon successful synthesis, with no partial or streaming data available.

How do I rotate my API key?

Generate a new key from your dashboard at any time. The new key becomes active immediately, and the old one stops working at once. This allows seamless rotation without downtime for your applications.

First request

curl

curl https://api.playf44.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.playf44.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.playf44.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

Authentication

Sound wave illustration for: Authentication

Authentication for the BlogSpeak API relies on a single Bearer token passed in the request header. Each account is issued one active API key, which provides access to all endpoints including speech synthesis and voice listings. This key must be included in every request as Authorization: Bearer tts-....

Bearer token format

Bearer token format

The API key is a string starting with tts- and must be included in every request header. It provides access to all endpoints, including speech synthesis and voice listings.

One active key per account

Each account holds exactly one active key. If you generate a new key, the previous one becomes invalid immediately. This ensures a single point of access for your billing and usage data.

Key rotation

You can issue a new key at any time from your dashboard. The old key stops working instantly, allowing you to rotate credentials without downtime. Use this if a key is exposed or if you manage multiple projects with separate access levels.

  • Generate new key instantly
  • Old key invalidates immediately
  • No downtime required

Endpoint Overview

Sound wave illustration for: Endpoint Overview

The API provides three core endpoints to manage audio generation and metadata. Use POST for synthesis, GET for listing available voices, and GET for model details. Each endpoint serves a distinct purpose in the audio creation workflow, from selecting a voice to retrieving the final audio file.

The POST endpoint handles the actual text-to-speech conversion. The GET /v1/voices endpoint returns a list of available voices, languages, formats, and pricing information. The GET /v1/models endpoint provides details about the available models.

POST /v1/audio/speech

This is the primary endpoint for generating audio. Send a JSON payload with your text, chosen voice, and desired format. The API processes the request and returns the audio file directly in the response body.

Ensure your text does not exceed 4096 characters. For longer content, split the text into smaller chunks and concatenate the resulting audio files. The API supports various output formats to suit different use cases, from web playback to high-quality archival.

  • Supports text input up to 4096 characters
  • Returns audio file directly in the response
  • Supports multiple output formats (mp3, wav, opus, flac, pcm)

GET /v1/voices

Retrieve a list of all available voices, their associated languages, supported formats, and pricing details. This endpoint helps you select the most appropriate voice for your content, whether it’s a warm narrator or a neutral newsreader.

The response includes details about each voice, such as its gender, tone, and the languages it can speak. You can use this information to build a dynamic voice selector in your application or to automate voice selection based on content type.

  • List of 8 voice presets
  • Supported languages and formats
  • Pricing information per voice

GET /v1/models

Get details about the available TTS models, including their names and capabilities. Currently, the API offers two models: tts-1 and tts-1-hd. Both use the same engine and pricing.

Use this endpoint to determine which model best suits your needs. Both models use the same engine and price. Both models support all 23 languages and 8 voice presets.

  • tts-1: Standard quality
  • tts-1-hd: High definition quality
  • Same pricing for both models

Request Parameters

Sound wave illustration for: Request Parameters

Configure generation parameters to control audio output. Select the model, define the input text, and choose a voice preset to match your content style.

Model selection

Choose between tts-1 and tts-1-hd. Both models use the same engine and price.

Input text limit

Pass up to 4096 characters per request. Longer texts require splitting into chunks and joining the resulting audio files.

Voice preset

Select one of eight presets. Voices include female options like Aria and Nova, and male options like Felix and Milo.

  • Aria: warm, clear narrator
  • Nova: bright, upbeat
  • Felix: energetic, promo
  • Atlas: neutral, news

Response Formats

Sound wave illustration for: Response Formats

Choose the audio format that best fits your workflow. MP3 is the standard for web playback and most podcast platforms, offering a balance of quality and file size.

For archival or editing purposes, lossless and uncompressed options are available. Select the format that matches your downstream processing needs.

MP3 output

MP3 is a widely supported format. It is ideal for embedding directly into blog posts or streaming in podcast players.

WAV and FLAC

WAV provides uncompressed PCM audio, while FLAC offers lossless compression. Both are suitable for high-fidelity editing or archival storage where file size is less critical than quality.

Opus and PCM

Opus provides efficient compression for modern applications. PCM returns raw audio data, useful for immediate processing in code without decoding overhead.

Rate Limits

Sound wave illustration for: Rate Limits

API usage is governed by strict rate limits to ensure stable generation for all users. Each API key allows up to 60 requests per minute, regardless of account funding status. This limit applies uniformly to all active keys on your account.

Before your first top-up, accounts are restricted to 10 requests per minute. This lower threshold helps manage initial load until you add USDT (TRC-20) credit. Once funded, your limit increases to the standard 60 requests per minute immediately.

60 requests per minute

Each API key is limited to 60 requests per minute. This cap applies uniformly, ensuring predictable performance whether you are generating short phrases or long-form content for your blog or podcast.

If you need higher throughput, you can issue a new API key from your dashboard. The old key stops working instantly, allowing you to scale your request capacity by managing multiple keys.

  • Limit applies per key, not per account.

10 requests before top-up

Accounts are restricted to 10 requests per minute before the first top-up. This restriction applies immediately upon creation, including accounts that start with free trial credit.

This lower limit helps manage server load during the initial phase. Once you top up with USTRC-20, the limit automatically increases to 60 requests per minute.

Queued processing

Requests are processed one after another. Under heavy load, a request may wait in a queue before synthesis begins. This sequential processing ensures stability but means generation times can vary slightly.

For most use cases, this is transparent. Typical speed is roughly real time, but short phrases may take a few seconds to generate. Plan your workflows accordingly, especially if generating many files in rapid succession.

Rate Limits

Error Handling

Sound wave illustration for: Error Handling

When synthesis fails, the API returns a non-200 HTTP status code alongside a JSON error body detailing the issue. Failed requests incur no charge, and the cost is returned in full to your balance.

Common errors include invalid API keys, exceeded character limits, or unsupported voice selections. The response includes a clear error code and message to help you debug your integration quickly.

HTTP status codes

The API uses standard HTTP status codes to indicate success or failure. A 200 OK response means the audio was generated successfully. Errors return 4xx codes for client issues (like invalid input) or 5xx for temporary server problems.

  • 400 Bad Request: Invalid parameters or malformed JSON.
  • 401 Unauthorized: Missing or invalid Bearer token.
  • 429 Too Many Requests: Rate limit exceeded.
  • 500 Internal Server Error: Temporary service issue.

JSON error bodies

Each error response includes a JSON object with an error code and a human-readable message. This structure allows you to programmatically identify the cause of the failure and handle it appropriately in your application logic.

  • error: A machine-readable error code.
  • message: A detailed description of the error.
  • type: The type of error (e.g., invalid_request).

Retry logic

For transient errors like 500 or 429, implement exponential backoff. Wait briefly before retrying, increasing the delay with each attempt. This prevents overwhelming the API during high load periods and ensures efficient resource usage.

  • Retry on 429: Wait and retry after a short delay.
  • Retry on 5xx: Use exponential backoff for temporary server issues.
  • Stop retrying on 4xx: Client errors like invalid input do not resolve with retries.

OpenAI SDK Compatibility

Sound wave illustration for: OpenAI SDK Compatibility

BlogSpeak’s audio endpoint mirrors the structure of OpenAI’s speech API. You can integrate our service into existing workflows by simply adjusting the API base URL. No custom client libraries or complex rewrites are required to start generating audio.

Base URL change

Point your SDK configuration to https://api.playf44.com/v1 instead of the default OpenAI endpoint. This small configuration shift allows your code to route requests to our infrastructure while maintaining standard authentication headers.

Standard request shape

The JSON body accepts the same fields: model, input, voice, response_format, and speed. Our engine interprets these standard parameters identically, ensuring predictable behavior for common text-to-speech operations.

Official SDKs work

The official OpenAI SDKs function correctly out of the box when configured with our base URL. You can use your existing code to generate MP3, WAV, or Opus files without writing custom HTTP clients.

OpenAI SDK Compatibility

cURL Example

Sound wave illustration for: cURL Example

The quickest way to verify your integration is a direct cURL command. This example demonstrates a standard synthesis request using the tts-1 model and the nova voice.

The API returns the audio file directly, allowing you to save it locally for immediate use in your blog or podcast workflow.

Basic request command

Use the following command to send text to the API. Replace YOUR_API_KEY with your actual key from the dashboard.

This command targets the correct base URL for playf44.com and specifies the speech endpoint.

  • -X POST method
  • -H "Authorization: Bearer YOUR_API_KEY" header
  • -H "Content-Type: application/json" header
  • -d data payload

JSON body structure

The request body contains the core parameters for synthesis. You can adjust the voice, speed, and output format within this structure.

Ensure your text does not exceed 4096 characters per request to avoid errors.

  • model: "tts-1" or "tts-1-hd"
  • input: "Your text here"
  • voice: "nova"
  • response_format: "mp3"

Audio file response

On success, the response body contains the raw audio bytes. You can redirect this output to a file using the -o flag.

If the synthesis fails, you will receive an HTTP error code and a JSON description of the issue.

Listen to the 8 voices

Each sample below was generated by this API from the same sentence. Mastering WordPress unlocks endless freelance opportunities for creative bloggers seeking financial independence.

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. Python documentation
  3. LibriTTS-R corpus (OpenSLR 141)
  4. curl documentation
  5. W3C WAI: making audio and video accessible
  6. Creative Commons Attribution 4.0
  7. IETF RFC 7845: Ogg Opus
  8. MDN: Web Speech API
  9. LibriVox public-domain audiobooks
  10. OpenAI API reference: create speech
  11. openai-node on GitHub
  12. Tether (USDT)

More on this site