Frequently asked questions
Which voice is best for luxury homes?
A: Atlas offers a neutral, news-style tone suitable for professional property tours, while Orion provides a deep, documentary feel for high-end listings. Aria delivers a warm, clear narration ideal for elegant walkthroughs. Select the voice that matches your brand’s specific aesthetic and target audience demographics. Note that any voice can speak any supported language, so choice depends on your desired tone rather than a single "best" option.
Can any voice speak any language?
Yes, all eight voices can speak all 23 supported languages. You can assign any voice to any language, such as using Aria for Spanish or Felix for German. If the language is not set, the system guesses it based on the script, defaulting to English for Latin text.
Is there a cost difference between voices?
No, there is no cost difference between voices. All voices use the same underlying engine. Pricing is based solely on character count: $10 per 1 million characters. The model choice (tts-1 or tts-1-hd) determines the price, not the selected voice preset.
What is the difference between tts-1 and tts-1-hd?
Both models use the same engine and have the same price of $10 per 1 million characters. The distinction is purely in audio quality, with tts-1-hd offering higher fidelity. Both support the same voices, formats, and speed settings within the API request structure.
Can I clone my own voice?
No, custom voice cloning is not offered. You cannot upload your own voice or design a new one. You must choose from the eight available presets: Aria, Nova, Luna, Sage, Felix, Milo, Atlas, and Orion, which are based on licensed recordings from LibriTTS-R and LibriVox.
Do voices sound natural?
The voices are based on licensed recordings from LibriTTS-R and LibriVox public-domain audiobook readings. They provide clear, natural-sounding speech suitable for property descriptions. However, every generated file carries an inaudible watermark marking it as synthetic speech, which distinguishes it from human-recorded audio.
Is there a watermark in audio?
Yes, every generated audio file carries an inaudible watermark that marks it as synthetic speech. This applies to all voices and languages. The watermark is embedded within the audio data and is not audible during playback but indicates the file is AI-generated.
Can I adjust voice speed?
Yes, you can adjust the speed between 0.25 and 4.0. This allows you to speed up property tours or slow down for detailed explanations. Adjustments are made via the speed field in the API request, affecting the generation time and final audio duration accordingly.
Are voices gender-specific?
The presets are categorized as female (Aria, Nova, Luna, Sage) or male (Felix, Milo, Atlas, Orion). However, any voice can speak any of the 23 supported languages. You can mix genders in a single tour by calling the API multiple times with different voice parameters.
Do voices work in all languages?
All eight voices work in all 23 supported languages, including Arabic, Chinese, English, German, Spanish, and Japanese. Unsupported languages like Vietnamese or Thai are not available. You must specify the language code if using Latin script for non-English languages to ensure correct pronunciation.
Is audio stored after generation?
No, generated audio is returned in the response and not stored by the API. Input text is used only to produce the audio and is not used for training. Only request metadata, such as time, characters, and status, is kept for billing purposes.
Can I use audio commercially?
The voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain readings. You can use the generated audio for commercial property tours. Ensure you comply with the specific license requirements of the source material if distributing the audio publicly.
Do I need to download audio?
The API returns the audio file directly in the response. You do not need to download it from a separate storage bucket. You can save the returned file to your local system or upload it directly to your virtual tour platform using the response data.
Is there a limit on voices?
There are exactly eight voice presets available: Aria, Nova, Luna, Sage, Felix, Milo, Atlas, and Orion. You can use any of these voices in your requests. There is no limit on how many times you can switch between them within a single tour or across different tours.
Can I use voices for ads?
Yes, you can use the generated audio for advertisements. The voices are suitable for promotional content, such as real estate ads or social media clips. Since the voices are based on licensed recordings, ensure you adhere to the CC BY 4.0 license terms for commercial use.
Do voices support punctuation?
Yes, voices naturally interpret punctuation to create pauses and intonation. You can use standard punctuation in your input text to control the flow of the narration. This helps create a more natural listening experience for property descriptions without needing complex SSML tags, which are not supported.
Is there a free trial?
Yes, signing in with Google provides a $0.50 free trial, valid for 7 days, which covers about 50,000 characters. Accounts created with email and password start without this trial credit. Failed requests are free, so you can test without charge if synthesis fails.
How do I change voices?
Change the voice by updating the voice field in your API request JSON. You can switch between any of the eight presets for each request. This allows you to vary the narration style within a single property tour or between different properties easily.
Can I mix voices in one tour?
Yes, you can mix voices by sending multiple API requests. Generate audio for different sections of your tour using different voices, then join the audio files. This allows for dynamic narration, such as using Atlas for facts and Aria for emotional highlights.
Is there a mobile SDK?
No, there is no official mobile SDK. You can use the official OpenAI SDKs by changing the base_url to https://api.play33bm.com/v1 or use plain HTTP requests. This allows you to integrate the TTS API into any mobile app or no-code platform that supports HTTP calls.
Female voices overview
The PropVoice API offers three distinct female voices, each engineered to enhance property tours with specific tonal qualities. These presets—Aria, Nova, and Luna—allow no-code creators to match the audio mood to the property type without external editing tools. All voices support 23 languages and are priced identically.
Aria: warm narrator
Aria delivers a clear, warm narration style ideal for standard residential listings. Her tone guides listeners through room descriptions smoothly, making her suitable for everyday walkthroughs where clarity matters more than dramatic effect.
Nova: bright upbeat
Nova provides a bright, energetic delivery perfect for promotional content or short highlight reels. Her upbeat cadence works well for commercial spaces or quick social media teasers where energy drives engagement.
Luna: soft storytelling
Luna offers a soft, storytelling voice that creates an intimate atmosphere. She is best suited for luxury estates or historical properties where a gentle, narrative tone enhances the sense of place and heritage.
Male voices overview
The male voices offer distinct tonal qualities suited for various property marketing needs, from energetic promotions to neutral reporting. Each preset is engineered to enhance specific aspects of a real estate narrative without requiring technical adjustments.
Felix: energetic promo
Felix delivers an upbeat, high-energy tone ideal for promotional listings. His delivery keeps viewers engaged during quick property highlights or sales announcements, ensuring the audio matches the excitement of a new listing.
Milo: friendly conversational
Milo provides a warm, approachable voice perfect for conversational walkthroughs. He sounds like a knowledgeable guide, making virtual tours feel personal and inviting for potential buyers browsing residential homes.
Atlas: neutral news
Atlas offers a clear, neutral delivery suitable for factual property data. His straightforward tone works well for listing details, pricing, and specifications, ensuring information is conveyed without emotional bias.
Orion voice details
Orion delivers a deep, documentary-style narration that adds gravitas to virtual property tours. Its low-frequency resonance creates an immersive atmosphere, making it ideal for high-end real estate content where tone matters as much as information.
The voice maintains consistent clarity even at slower speeds, ensuring every detail of a luxury estate is understood. It works well for guiding users through expansive properties without sounding artificial.
Deep documentary tone
Orion’s rich, low-pitched delivery mimics professional documentary narration. This tone adds a layer of sophistication to property tours, particularly for luxury estates where ambiance influences perception.
- Low-frequency resonance
- Documentary-style pacing
- Sophisticated audio profile
Suitable for luxury tours
For high-value listings, Orion’s authoritative yet calm presence enhances the perceived value of the property. It pairs well with slow-paced walkthroughs, allowing listeners to absorb visual details without distraction.
- Enhances perceived property value
- Complements slow-paced tours
- Ideal for luxury estates
High clarity
Despite its deep tone, Orion maintains exceptional clarity. Even when sped up to 2.0x for quick tours, the voice remains distinct and easy to follow, ensuring no key details are lost.
- Clear articulation at speed
- Easy to follow
- No distortion at high speeds
Choosing based on property type
To match property type with voice, select tones that align with audience expectations. Residential listings benefit from warm or bright narration, while commercial spaces require neutral clarity.
Luxury estates demand a deep, authoritative tone that conveys exclusivity. Choose voices carefully to enhance the perceived value of the listing without distracting from the visual tour.
Residential listings
Aria’s warm narration or Nova’s upbeat style works well for family homes. These voices create an inviting atmosphere for buyers touring their future living spaces.
- Aria: warm, clear narrator for family homes
- Nova: bright, upbeat for energetic listings
Commercial spaces
Atlas provides a neutral, news-style delivery ideal for office buildings or retail spaces. His tone ensures information is delivered clearly without emotional bias.
Milo’s conversational style can also work for mixed-use developments, adding a friendly touch to complex property details.
- Atlas: neutral, news-style for office buildings
- Milo: friendly, conversational for mixed-use
Luxury estates
Orion’s deep, documentary tone suits high-end properties. His voice conveys authority and sophistication, enhancing the prestige of luxury estates.
Use Orion for detailed descriptions of premium features, ensuring the audio reflects the exclusivity of the property.
- Orion: deep, documentary tone for luxury
- Sage: calm, instructional for detailed guides
Voice selection by language
PropVoice supports 23 languages, covering major markets for global real estate. Whether you are listing a apartment in Berlin or a villa in Tokyo, the API handles the audio generation without needing separate voice models for each region.
23 languages supported
The API covers Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, and Turkish. This list includes all primary languages for international property tours, ensuring your virtual guides sound native to local audiences.
Any voice speaks any language
Unlike systems that lock voices to specific languages, PropVoice allows any of the 8 presets to speak any of the 23 supported languages. You can use the warm narrator voice for a Spanish listing or the calm instructional voice for a German commercial space, maintaining consistent brand tone across markets.
- Mix voices freely across different language tours.
Latin text defaults to English
If you do not specify a language code, the system guesses the language based on the script. Latin-script text defaults to English. For non-English Latin-script text (e.g., French or German), you must set the language parameter explicitly to ensure accurate pronunciation and avoid English pronunciation artifacts.
Audio format considerations
Selecting the correct audio format ensures your virtual property tours load quickly while maintaining professional clarity. Each format serves a distinct purpose in the deployment workflow, from immediate web playback to long-term archival storage.
The API supports mp3, wav, opus, flac, and pcm. Choosing wisely affects bandwidth costs and listener experience without changing the voice quality or price.
MP3 for web
MP3 is the standard for web delivery due to its small file size and universal browser support. It provides sufficient clarity for voice guidance during virtual walkthroughs, ensuring fast load times even on slower connections.
This format is ideal for embedding audio directly into no-code website builders or mobile apps where bandwidth efficiency matters most.
- Best for mobile virtual tours
- Small file size reduces loading time
- Universal playback support
WAV for high quality
WAV delivers uncompressed, high-fidelity audio, preserving every nuance of the synthetic voice. Use this format when audio quality is prioritized over file size, such as in high-end marketing presentations or local playback.
Ideal for users who want to ensure the voice sounds exactly as generated, without any compression artifacts.
- Uncompressed audio fidelity
- Ideal for marketing materials
- Larger file size
FLAC for archives
FLAC offers lossless compression, reducing file size by approximately 50–70% compared to WAV while retaining full audio quality. This makes it perfect for storing generated voiceovers for future reuse without re-generating.
Use FLAC when you need to preserve the original quality for editing or long-term storage without consuming excessive disk space.
- Lossless compression
- Ideal for archival storage
- Preserves original quality
Pricing impact of voices
Every voice preset costs the same: $10 per 1 million characters. You can switch between Aria, Felix, or Orion without any price difference. The rate is $0.01 per 1,000 characters, including spaces and punctuation. Failed requests are free, so your budget stays predictable.
Same price for all voices
All eight voices are priced identically. Whether you use Luna for soft storytelling or Atlas for neutral news, the cost per character remains constant. This allows you to A/B test different tones for your virtual property tours without worrying about variable billing.
- Flat rate: $10 per 1 million characters
- No voice-specific premiums
- Spaces and punctuation count as characters
tts-1 and tts-1-hd
The API offers two model versions: tts-1 and tts-1-hd. Both use the same underlying engine and incur the same charge. You can choose the HD model for higher fidelity without paying extra, or stick with the standard model.
- Both models cost $10 per 1M characters
- HD offers higher audio fidelity
- Standard model may process slightly faster
No extra cost for HD
Upgrading to high-definition audio does not increase your bill. The price per character is identical regardless of the response_format or model version selected. This simplicity helps no-code creators integrate TTS into their property management workflows without calculating complex tiered costs.

Technical requirements
For virtual property tours, input text must not exceed 4,096 characters per request. Longer descriptions require splitting text into parts and joining the resulting audio files. This limit ensures stable processing within the API’s standard workflow.
4096 character limit
Each API call handles up to 4,096 characters, including spaces and punctuation. For longer property descriptions, split the text into chunks and concatenate the generated audio files to create a seamless guided tour experience.
Speed control 0.0.25–4.0
Adjust playback speed between 0.25 and 4.0 to match the pacing of your video walkthrough. Faster speeds reduce generation time, while slower speeds enhance clarity for complex property details or architectural terminology.
API key authentication
Authenticate requests using a Bearer token in the Authorization header. Each account holds one active key; issuing a new key immediately invalidates the previous one. Use this key to access the API base at https://api.play33bm.com/v1.
Quality and realism
The synthetic speech is built from licensed recordings sourced from LibriTTS-R (CC BY 4.0) and public-domain LibriVox audiobook readings. This foundation ensures clear, natural-enough audio suitable for guiding virtual property tours without sounding like a rigid robot.
Every generated file carries an inaudible watermark that marks it as synthetic speech. This allows platforms to identify AI-generated audio automatically.
Licensed recordings
Our voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and public-domain LibriVox audiobook readings. This provides a natural baseline for clarity and tone across all 8 presets.
Inaudible watermark
Each audio file includes a subtle, inaudible watermark identifying it as synthetic. This helps platforms verify the origin of your property tour audio.
Synthetic speech detection
Modern AI detectors can identify these watermarks. When sharing tours on third-party platforms, this transparency may help avoid confusion about whether the voice is human or AI-generated.
Common use cases
Text-to-speech APIs enable real estate professionals to convert property descriptions into natural-sounding audio files, creating immersive guided walkthroughs without hiring voice actors. This workflow is ideal for no-code creators building virtual tours who need consistent, high-quality narration at scale.
By leveraging the API’s character limits and speed controls, users can generate audio files that match the pacing of video tours, ensuring a professional and engaging experience for potential buyers.
Guided virtual tours
Use the API to generate narrated audio for virtual property tours. Select voices like Aria for warm narration or Felix for energetic promos to match the property’s vibe. Audio files can be synced with video tours for a seamless experience.
With support for 23 languages, you can create localized tours for international buyers without additional voice talent costs. Each request generates an audio file up to 4096 characters, which can be split and joined for longer properties.
- - Generate narration for video tours
- - Localize tours for global buyers
- - Sync audio with video timelines
Property listing audio
Convert written property listings into audio files for websites or apps. Use voices like Luna for soft storytelling or Milo for friendly, conversational tones. This adds accessibility and engagement to standard text descriptions.
Audio files are returned directly in the API response, allowing easy integration into no-code platforms. The inaudible watermark marks each file as synthetic speech, maintaining transparency.
- - Enhance text listings with audio
- - Improve accessibility for visually impaired buyers
- - Maintain transparency with synthetic speech markers
Multi-language tours
Support 23 languages with any voice, enabling multilingual property tours. For non-English Latin-script text, set the language parameter to ensure accurate pronunciation. This is ideal for international real estate markets.
Use the API’s speed control (0.25–4.0) to adjust narration pace for different languages or audience preferences. Audio formats like MP3 or WAV can be chosen based on web or quality needs.
- - Support 23 languages with any voice
- - Set language parameter for accuracy
- - Adjust speed for audience preferences
Integration tips
Integration is straightforward for no-code developers using standard HTTP libraries or the OpenAI SDK. By updating the base URL and providing your API key, you can generate audio files for property tours with minimal code changes.
Use OpenAI SDK
The same request shape as the OpenAI speech endpoint allows you to reuse existing client libraries. This simplifies integration for teams already familiar with standard TTS workflows.
- Reuse existing OpenAI SDK code
- Standard JSON request body
- Familiar parameter structure
Change base URL
Update the SDK configuration to point to https://api.play33bm.com/v1. This directs requests to our infrastructure while maintaining compatibility with standard TTS parameters.
- Base URL: https://api.play33bm.com/v1
- Keep standard headers
- Use your API key
Handle errors gracefully
Implement retry logic for rate limits (60 requests per minute). Failed synthesis requests are free, so you can retry without incurring extra charges for the same text.
- Retry on 429 Too Many Requests
- Check response status codes
- Free retries on failure
Limitations to note
While PropVoice delivers natural-sounding audio for real estate tours, it lacks advanced voice customization features. You cannot clone specific human voices or upload custom recordings, meaning all audio is generated from the eight standard presets. This ensures consistency but removes the ability to create a unique brand voice.
The API also does not support SSML tags for fine-grained control over pronunciation or pacing, nor does it offer streaming audio. You receive the complete file only after generation finishes, which may impact real-time interactive tour experiences.
No voice cloning
Custom voice cloning is not offered. You cannot upload a sample to replicate a specific agent’s voice or create a proprietary brand sound. All generations use the eight available presets (Aria, Nova, Luna, Sage, Felix, Milo, Atlas, Orion).
No SSML support
The API does not support SSML (Speech Synthesis Markup Language). You cannot inject tags to control pauses, emphasis, or pronunciation of specific words. Pacing is controlled solely via the global speed parameter (0.25–4.0).
No streaming audio
Audio is returned as a complete file, not streamed in real-time. This means you must wait for the full synthesis to finish before playing the audio, which is suitable for pre-recorded tours but not live interactive walkthroughs.
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.
