Frequently asked questions
How do I get an API key?
Sign up at https://play33bm.com/get-api-key/ using Google or email. Your key appears immediately on the dashboard. One active key is issued per account. You can generate a new key at any time, and the old one stops instantly. No developer account is required beyond standard registration.
What is the base URL?
The API base URL is https://api.play33bm.com/v1. Use this for all requests. The speech endpoint is POST https://api.play33bm.com/v1/audio/speech. Voices and models are available at GET /v1/voices and GET /v1/models respectively. Ensure your requests target this specific domain for proper authentication and routing.
Can I use OpenAI SDK?
Yes. The request shape matches the OpenAI speech endpoint. Change the SDK’s base_url to https://api.play33bm.com/v1. Use your PropVoice key for authentication. The official OpenAI SDKs work seamlessly with this configuration, allowing you to integrate TTS into your existing codebase without learning a new library.
What models are available?
Two models exist: tts-1 and tts-1-hd. Both use the same engine and price. Select either based on your preference. The API treats them identically for speech synthesis. There are no other model variants or tiers available in the current system.
Which voices can I use?
Eight presets are available: aria, nova, luna, sage, felix, milo, atlas, and orion. Each voice supports all 23 languages. Select a voice by name in the request. Voices are based on licensed recordings from LibriTTS-R and LibriVox. No custom cloning or emotion tags are supported.
What response formats are supported?
Supported formats are mp3, wav, opus, flac, and pcm. Specify the format in the response_format field. All formats include an inaudible watermark marking the audio as synthetic. Choose mp3 for general use or wav/flac for higher fidelity in property tours.
What are the rate limits?
Limits are 60 requests per minute per key. New accounts start with 10 requests per minute until the first top-up. Requests are processed sequentially, so load may cause queuing. Exceeding limits may result in errors. Monitor your usage via the dashboard to avoid interruptions.
How do I handle errors?
Failed requests are free; charges are returned in full. Check the response status code for details. Ensure your text is under 4096 characters and your key is valid. Use the dashboard to view recent usage and error logs. Support is available via email if issues persist.
Is there an SSML option?
No, SSML is not supported. The API accepts plain text input only. You cannot use SSML tags for pronunciation or pauses. Control pacing by adjusting the speed parameter between 0.25 and 4.0. For precise control, split your text into smaller segments.
Can I stream audio?
No streaming or WebSocket support exists. Audio is returned as a complete file in the response. You cannot receive partial audio chunks. This is suitable for static property tours but not for real-time interactive applications. Wait for the full file before playing.
How do I split long texts?
Split text into parts of up to 4096 characters. Generate audio for each part separately. Join the resulting audio files client-side. This ensures each request stays within limits. Use standard string splitting logic in your code to handle longer property descriptions.
What is the character limit?
The limit is 4096 characters per request. Spaces and punctuation count toward this total. Longer texts must be split manually. Each character costs $0.00001. Plan your splits to minimize API calls while maintaining audio continuity for your virtual tours.
Do I need a developer account?
No specialized developer account is needed. Sign up via Google or email at the registration page. Accounts are per site. You receive an API key and dashboard access immediately. No minimum spend or monthly plans are required to start using the API.
How do I change my API key?
Issue a new key from your dashboard at any time. The old key stops working immediately upon generating the new one. Only one key is active per account. Update your applications with the new key to ensure continued service without downtime.
First request
curl
curl https://api.play33bm.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.mp3Python (openai)
from openai import OpenAI
client = OpenAI(base_url="https://api.play33bm.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.play33bm.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 | |
|---|---|
aria | female |
nova | female |
luna | female |
sage | female |
felix | male |
milo | male |
atlas | male |
orion | male |
| language | |
|---|---|
ar | العربية |
zh | 中文 |
da | Dansk |
nl | Nederlands |
en | English |
fi | Suomi |
fr | Français |
de | Deutsch |
el | Ελληνικά |
he | עברית |
hi | हिन्दी |
it | Italiano |
ja | 日本語 |
ko | 한국어 |
ms | Bahasa Melayu |
no | Norsk |
pl | Polski |
pt | Português |
ru | Русский |
es | Español |
sw | Kiswahili |
sv | Svenska |
tr | Türkçe |
Get your API key
Accessing the PropVoice API requires a valid authentication key. Sign up via the /get-api-key/ page using Google or an email account. Once registered, you can generate a key immediately. This key grants access to the API endpoints for generating speech files.
The dashboard displays your current key, balance, and usage statistics. Use this key in the Authorization header for all requests. Keys are unique per account and can be regenerated at any time, instantly invalidating the previous one.
Sign up on the site
Visit the /get-api-key/ page to create an account. You can sign up using Google for instant access or provide an email and password. Accounts are created per site, ensuring your credentials are specific to this storefront.
- Google sign-in for immediate access
- Email and password option available
- Instant account creation
Retrieve key from dashboard
After signing in, navigate to the dashboard to view your active API key. The dashboard also shows your balance, trial status, and character usage. Copy the key string starting with 'tts-' for use in your requests.
You can issue a new key at any time. When a new key is generated, the old one stops working immediately, ensuring secure access control without manual deactivation.
- View active key and balance
- Issue new key instantly
- Old key invalidates immediately
Use Bearer token auth
Include your API key in the Authorization header of every request. Use the format: 'Authorization: Bearer tts-...'. This standard authentication method ensures secure communication with the API endpoints.
The key is required for all speech generation requests. If the key is missing or invalid, the API will return an authentication error. Keep your key secure and do not expose it in public client-side code.
- Header: Authorization: Bearer tts-...
- Required for all requests
- Invalidate old keys instantly
First request with curl
Use the standard POST endpoint to generate audio files from your property descriptions. The API accepts text input and returns synthesized speech in your chosen format, enabling quick integration into no-code virtual tour builders without complex setup.
Authentication requires a Bearer token sent in the header. Each request processes text up to 4096 characters, returning a direct audio file in the response body.
POST endpoint URL
Send your request to https://api.play33bm.com/v1/audio/speech. This endpoint handles all synthesis tasks and returns the audio file directly in the response body.
JSON body structure
The request body must be valid JSON containing the model, input text, voice, and response_format. The input field accepts up to 4096 characters of property description text.
Include optional parameters like speed (0.25–4.0) to adjust narration pace. Ensure the Authorization header contains your Bearer token for validation.
- model: "tts-1" or "tts-1-hd"
- input: string (max 4096 chars)
- voice: one of 8 presets
- response_format: mp3, wav, opus, flac, or pcm
Response format
On success, the response contains the generated audio file directly in the body. The format matches your specified response_format parameter, ready for playback in web players or download.
If the request fails, you receive a JSON error object with details. Failed requests incur no charge, and the balance is returned in full.
Using OpenAI SDK
The OpenAI Python and Node.js SDKs work directly with PropVoice by overriding the API base URL. Point the client to https://api.play33bm.com/v1 instead of the default OpenAI endpoint. This allows you to use familiar SDK methods without rewriting your integration logic.
Since the request shape matches the OpenAI speech endpoint exactly, you pass your PropVoice API key via the Authorization: Bearer header. The SDK handles authentication and JSON serialization, letting you focus on generating audio for property tours using standard library functions.
Change base_url
Initialize your SDK client and set the base_url to https://api.play33bm.com/v1. This directs all subsequent API calls to the PropVoice infrastructure while maintaining the standard OpenAI request structure.
- Set
base_url = 'https://api.play33bm.com/v1' - Keep the standard SDK initialization pattern
- Ensure the key starts with
tts-
Same request shape
Use the identical client.audio.speech.create() method. Pass parameters like model, input, voice, and response_format exactly as you would for OpenAI. The engine processes these fields identically.
- Use
client.audio.speech.create() - Pass standard fields:
model,input,voice - No special wrapper functions needed
Official SDKs work
The official OpenAI SDKs (Python, Node.js) are fully compatible. You do not need a proprietary PropVoice client. Simply update the base URL and inject your unique tts-... API key for authentication.
Available parameters
The API accepts two model identifiers, tts-1 and tts-1-hd, which utilize the same underlying engine and pricing structure. Both handle the standard request shape, allowing you to choose based on your quality preference without changing your integration logic. Text input is limited to 4096 characters per request, requiring simple splitting for longer property descriptions.
You can adjust the playback speed between 0.25 and 4.0, enabling faster walkthroughs for quick overviews or slower narration for detailed room features. The voice parameter selects one of eight presets, such as atlas for neutral news-style delivery or aria for warm narration, ensuring the audio matches the tone of your virtual tour.
Model: tts-1 or tts-1-hd
Both models use the same engine and cost $0.01 per 1,000 characters. The choice does not affect pricing or voice quality, so tts-1 is sufficient for most real estate applications.
The maximum input length is 4096 characters. For longer property listings, split the text into parts and join the resulting audio files sequentially.
- Model:
tts-1ortts-1-hd - Same engine and price
- 4096 character limit per request
Voice presets list
Select from eight licensed voices based on your brand voice. Male options include atlas for neutral news or felix for energetic promos. Female options include aria for warm narration or nova for upbeat tours.
Each voice is based on licensed recordings from LibriTTS-R. All generated files carry an inaudible synthetic watermark. Choose voices that match the professional tone of your virtual property tours.
- Female: aria, nova, luna, sage
- Male: felix, milo, atlas, orion
- Licensed from LibriTTS-R
Speed range 0.25–4.0
Adjust speed from 0.25x to 4.0x to control narration pacing. Slower speeds aid clarity for complex property details, while faster speeds work for quick overviews of multiple listings.
The default speed provides a natural listening experience. Requests are processed sequentially, so higher speeds may reduce generation time.
- Range: 0.25 to 4.0
- Controls playback speed
- No impact on queue order
Response formats
The API returns raw audio data directly in the HTTP response body, eliminating the need for separate download requests. You specify the desired output format via the response_format parameter in your JSON payload. Supported formats include MP3, WAV, Opus, FLAC, and PCM, allowing flexibility for different playback environments and storage requirements.
Each generated file is returned as a binary stream. The response includes standard HTTP headers indicating content type, ensuring your application can save the file correctly. This direct delivery method keeps the workflow simple for no-code builders integrating audio into virtual property tours.
MP3, WAV, Opus, FLAC, PCM
Select your preferred format using the response_format field. MP3 offers high compression for smaller file sizes, while WAV provides uncompressed quality. Opus and FLAC offer efficient compression with high fidelity, and PCM provides raw, uncompressed audio data. Choose based on your storage constraints and playback device capabilities.
- MP3: Compressed, widely compatible
- WAV: Uncompressed, high quality
- Opus: Efficient compression
- FLAC: Lossless compression
- PCM: Raw audio data
Audio file in response
The synthesized audio is delivered directly in the response body. There is no intermediate storage step; the audio is generated and sent immediately upon successful synthesis. This ensures low latency for short phrases and straightforward integration into your application logic.
No streaming support
The API does not support streaming or WebSocket connections for real-time audio playback. The entire audio file is returned after synthesis completes. For longer property descriptions, ensure your application handles the complete binary response before attempting to play or save the audio file.

Rate limits and errors
Your account key allows 60 synthesis requests per minute. This limit applies to the API key, not your IP address. If you exceed this threshold, subsequent requests will be rejected with a rate limit error until the window resets.
60 requests per minute
Each key supports a maximum of 60 requests per minute. Before your first top-up, this limit is lower at 10 requests per minute. Once you add credit via USDT, the standard 60 RPM limit applies immediately.
- Standard limit: 60 requests/minute
- Pre-top-up limit: 10 requests/minute
- Limit is per API key
Queue processing under load
Requests are processed sequentially. If you send multiple tour audio files simultaneously, each request waits its turn in line. This means generation time depends on the queue length and text length, so long texts may take longer during peak usage.
Common error codes
Errors typically indicate invalid parameters, authentication failures, or rate limits. Failed requests are free, and the charge is returned in full if synthesis fails. Ensure your text stays within the 4096 character limit per request to avoid truncation errors.
- 429: Rate limit exceeded
- 401: Invalid or expired API key
- 400: Invalid request parameters or text too long
Voices and languages
The API supports 23 languages, allowing any voice preset to speak any language in the list. If the language field is omitted, the system guesses the language from the script, defaulting to English for Latin text. For non-English Latin-script descriptions, explicitly set the language parameter to ensure accurate pronunciation.
GET /v1/voices endpoint
Use GET https://api.play33bm.com/v1/voices to retrieve the full list of available voices, supported languages, and formats. This endpoint helps you verify which voice presets are active and their associated pricing before making synthesis requests.
8 voice presets
Select from eight licensed voice presets based on LibriTTS-R and LibriVox recordings. Choose female voices like Aria (warm narrator) or Nova (upbeat), or male voices like Felix (energetic) or Atlas (neutral news). Each voice carries an inaudible watermark marking it as 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
23 supported languages
Supported languages include Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, and Turkish. Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, Bengali, and others are not supported.
Example use cases
The API converts written property listings into audio guides, allowing no-code developers to embed voiceovers directly into static tour pages. By splitting long descriptions into 4096-character chunks, you generate seamless audio segments that play sequentially, ensuring detailed features are heard without overwhelming listeners.
For virtual walkthroughs, pair the tts-1-hd model with warm voices like Aria or Luna to create immersive guided tours. The synthetic watermark marks the audio as synthetic, while the natural pacing enhances engagement for remote buyers.
Property description audio
Convert bullet-point listings into natural speech. Use the input field with property details, and set voice to Sage for a calm, instructional tone that explains floor plans clearly.
Split lengthy descriptions into parts under 4096 characters. The API returns separate audio files that you join client-side, maintaining high quality without hitting length limits.
- Ideal for detailed feature lists
- Maintains clarity with shorter segments
- Works with any static website builder
Virtual tour narration
Use the speed parameter (0.25–4.0) to match the visual pacing of your video tours. Faster speeds work for quick overviews, while slower rates suit detailed architectural highlights.
Since requests process sequentially, plan your tour structure to minimize wait times during generation. Use the OpenAI SDK by pointing base_url to https://api.play33bm.com/v1 for easy integration.
- Adjust speed for tour pacing
- Sequential processing requires planning
- No streaming, so pre-generate assets
Multi-language support
Support global buyers by generating audio in 23 languages. Set the language parameter explicitly for non-English Latin script text to ensure correct pronunciation, as the API defaults to English for Latin characters.
Voices like Felix or Orion can deliver energetic or deep narration in Spanish, French, or German, allowing you to create localized tours without hiring multiple voice actors.
- 23 languages supported
- Explicit language setting recommended
- No custom voice cloning needed
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Explore every room in stunning detail with our easy virtual tours designed for no-code creators.
23 languages
Any voice speaks any of these languages.
Sources and further reading
- Wikipedia
- W3C WAI: making audio and video accessible
- LibriTTS-R corpus (OpenSLR 141)
- Tether (USDT)
- Python documentation
- TRON developer hub: TRC-20
- ISO 639 language codes
- LibriTTS-R paper (arXiv 2305.18802)
- FFmpeg documentation
- OpenAI API reference: create speech
- LibriVox public-domain audiobooks
- IETF RFC 7845: Ogg Opus
