Frequently asked questions
Who runs HomeEcho?
HomeEcho is an independent storefront for a cloud text-to-speech API. Accounts are managed per site. Accounts are managed per site, and we do not have a larger corporate parent. Our structure is designed for direct developer access without intermediate layers or enterprise overhead.
Are the voices real people?
The voices are based on licensed recordings from LibriTTS-R, which uses readings from LibriVox public-domain audiobooks. While the source material comes from human speakers, the output is synthetic. Every generated file carries an inaudible watermark that marks it as synthetic speech, distinguishing it from direct human recordings.
Do you use user text for training?
No. Input text is used only to produce the audio response and is not used for training models. Generated audio is returned in the response and not stored. Only request metadata, such as time, character count, and status, is kept for billing purposes and audit trails.
How are voices licensed?
Our eight voice presets are based on the LibriTTS-R dataset, licensed under CC BY 4.0. This data originates from LibriVox public-domain audiobook readings. This licensing model allows us to provide consistent, deterministic audio output for smart home and IoT developers without complex third-party voice rights management.
Where is the data stored?
Generated audio is returned in the response and is not stored by our servers. We retain request metadata, including timestamp, character count, and status, strictly for billing and usage tracking. Input text is processed to generate audio and is not retained for long-term storage or analysis.
Can I contact you by phone?
No. We do not offer a phone number or live chat support. All support is handled via email only. This approach ensures that developer inquiries are documented and addressed systematically without the noise of voice calls, aligning with our focus on API reliability and asynchronous communication.
Is HomeEcho part of OpenAI?
No. We are not affiliated with OpenAI, despite our API endpoint and request shape being compatible with the OpenAI speech endpoint. We use the same JSON structure but operate independently. Our base URL is https://api.playbk3.com/v1, and we maintain our own keys, pricing, and voice models.
Do you have an office?
We do not advertise a physical office location. As a cloud API provider, our operations are distributed. We do not offer on-premise installations or data-center choices for clients. Our infrastructure is managed entirely in the cloud, ensuring consistent access to the TTS API from any location globally.
How do I report a bug?
You can report bugs or issues via email support. Since we do not have phone support or live chat, email is the primary channel for technical communication. Please include your request ID and relevant metadata to help us diagnose issues with your text-to-speech synthesis requests efficiently.
Is the editorial team technical?
Yes. Our team understands API integration, rate limits, and audio formats like MP3, WAV, OPUS, FLAC, and PCM. We design our documentation and endpoints to be developer-friendly, ensuring that indie developers and IoT engineers can integrate our TTS service into smart home devices with minimal friction.
Can I use the voices commercially?
Yes. The underlying LibriTTS-R data is CC BY 4.0, and our service allows commercial use of the generated audio. You pay for character usage, and there are no restrictions on using the output for voice assistants, device notifications, or other commercial IoT applications as long as you have sufficient balance.
Do you offer refunds?
Failed requests are free, and the charge is returned in full if synthesis fails. We do not offer general refunds for successful requests. Your balance does not expire, so you can use your credit whenever needed. Top-ups via USDT on TRON are processed automatically within minutes.
Is the license perpetual?
Yes, your account balance does not expire. Once you top up via USDT, the credit remains available indefinitely. There are no monthly plans or subscriptions, so you retain access to your funds and can generate audio at any time without worrying about service renewal or license expiration.
Can I remove the watermark?
No. Every generated file carries an inaudible watermark that marks it as synthetic speech. This is a standard feature of our output to identify the audio as AI-generated. There is no option to remove this watermark, and it does not affect the audio quality or playback.
Do you hire developers?
We are an independent storefront and do not currently advertise job openings or hiring processes on this site. Our focus is on providing a reliable text-to-speech API for IoT developers. If you are interested in our service, you can sign up for an account and start using the API immediately.
What is the mission of HomeEcho?
HomeEcho provides reliable text-to-speech synthesis for smart home and IoT developers. We ensure consistent audio output without external dependencies. Our API handles voice generation at scale, ensuring deterministic results for voice assistants and device notifications, allowing developers to integrate high-quality audio into their hardware ecosystems efficiently.
Editorial Team
HomeEcho operates as an independent storefront for a unified text-to-speech API, serving smart home and IoT developers who require deterministic audio output. Our development team maintains full control over the API infrastructure, ensuring consistent performance for voice assistants and device notifications without external dependencies.
Independent Development
We build and manage the entire stack from the ground up. This includes the endpoint at https://api.playbk3.com/v1, the voice selection logic, and the billing system. By owning the full pipeline, we ensure that every request, from the initial JSON body to the final audio file, meets the specific needs of IoT integrations.
Technical Expertise
Our team specializes in low-latency audio synthesis for embedded systems. We understand the constraints of smart home hardware, where bandwidth and processing power are limited. Our API returns standard formats like MP3 and WAV, designed for easy playback on a wide range of devices without requiring complex decoding libraries.
No Corporate Backing
HomeEcho is not affiliated with major cloud providers or voice technology giants. We do not rely on the infrastructure of OpenAI, Google, or Amazon for our core voice engine. This independence allows us to offer a stable, predictable service tailored specifically for IoT developers who need reliability without corporate overhead.
Voice Licensing
HomeEcho’s voice engine relies on licensed recordings to ensure consistent audio quality for IoT applications. We utilize the LibriTTS-R dataset, which provides a reliable foundation for our eight preset voices. This approach allows us to maintain deterministic results for voice assistants and device notifications without the variability of synthetic generation alone.
LibriTTS-R Source
The core of our synthesis engine is the LibriTTS-R dataset. This resource offers high-quality audio samples that form the basis for our voice presets, ensuring clear and natural speech output for smart home devices.
CC BY 4.0 License
Our voice data is licensed under Creative Commons Attribution 4.0 International. This license permits the use of the recordings for our API services, allowing us to distribute synthesized audio to developers without additional licensing hurdles for end-user devices.
LibriVox Recordings
The source material consists of readings from LibriVox. These volunteer-read recordings provide the diverse vocal characteristics found in our eight presets, ranging from warm narrators to energetic promotional voices.
Data Policy
HomeEcho processes input text exclusively for the purpose of generating the corresponding audio file. We do not use your input text for training our models or for any secondary purposes beyond immediate synthesis. This ensures that your voice assistant scripts and device notification texts remain private and are not repurposed by our infrastructure.
Input Text Usage
Text is sent to the engine, converted to audio, and returned to you. The raw input is not stored long-term, ensuring that your application’s unique content remains yours.
Audio Retention
Generated audio files are returned in the API response and are not permanently stored by HomeEcho. Once the response is delivered, the audio data is discarded from our active servers, meaning you retain full control over the generated files.
- Audio is returned in the response and not stored.
Billing Metadata
We retain request metadata, including timestamp, character count, and status, strictly for billing accuracy. This data allows us to verify charges against your account balance and process refunds for failed requests without storing the actual audio or input content.
- Time, characters, and status are kept for billing.
Technical Limits
HomeEcho delivers deterministic audio synthesis via standard HTTP POST requests. Each request generates a complete audio file, ensuring consistent output for IoT device notifications and voice assistant triggers without the complexity of managing partial streams. This approach guarantees that your smart home automation logic receives fully formed audio assets, simplifying integration and reducing latency variability in constrained environments.
No Streaming Support
The API does not support WebSocket or streaming endpoints. Audio is returned as a complete file in the response body, which aligns with typical IoT data transfer patterns where partial audio chunks are rarely required.
No Custom Cloning
Customers cannot upload custom voice samples or clone specific speakers. The system uses eight predefined voices derived from licensed recordings, ensuring consistent audio characteristics across all generated content without the need for additional training data or voice design workflows.
No SSML
The endpoint does not interpret SSML tags. Prosody and pacing are controlled solely through the speed parameter (0.25–4.0), allowing developers to adjust generation time without managing complex XML structures or phoneme dictionaries.

Contact Information
HomeEcho operates as an independent storefront for the HomeEcho TTS API. Our support structure is designed for asynchronous efficiency, prioritizing clear written communication over real-time interruptions. This approach ensures that technical queries from IoT developers are handled with precision and documented for future reference without the noise of live channels.
Email Support
All inquiries are managed via email. This channel allows our engineering team to provide detailed, accurate responses to technical questions regarding voice selection, rate limits, or integration issues. We prioritize thoroughness in written replies to ensure your implementation remains stable.
No Phone Number
We do not maintain a dedicated phone line for customer support. This decision reduces overhead and keeps our API pricing competitive for indie developers. You can reach us at any time via email, regardless of your local time zone.
No Live Chat
There is no live chat widget on this site. We avoid real-time chat to prevent fragmented communication and ensure that support interactions are searchable and traceable. For urgent matters, email remains the most reliable and efficient method of contact.
Affiliation Status
HomeEcho operates as an independent storefront for the same underlying text-to-speech API. We are not affiliated with OpenAI, despite our endpoint compatibility, nor are we connected to Google, Amazon, Microsoft, or ElevenLabs. Our infrastructure and billing are entirely separate from these brands.
Not Affiliated with OpenAI
Our API structure mirrors the OpenAI speech endpoint, allowing developers to swap the base_url to https://api.playbk3.com/v1 and use the official OpenAI SDKs seamlessly. However, we are a distinct entity. We do not use OpenAI’s models, servers, or brand identity. Our keys and billing are managed independently.
Not Affiliated with Google
While we accept Google accounts for sign-in convenience, this is a standard authentication method and does not imply corporate ownership or partnership with Google. Our voice synthesis engine, data processing, and pricing model are entirely independent of Google Cloud Text-to-Speech services.
Independent Storefront
We are an independent team providing reliable TTS synthesis for IoT developers. Our site is a standalone storefront. We do not act as a reseller or subsidiary of any major tech brand. All voice assets, including the LibriTTS-R sources, are managed directly by our team for consistent smart home audio output.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. This allows you to test voice presets like Aria or Felix without managing complex billing setups.
23 languages
Any voice speaks any of these languages.
Sources and further reading
- Wikipedia
- Tether (USDT)
- IETF RFC 7845: Ogg Opus
- openai-node on GitHub
- LibriTTS-R corpus (OpenSLR 141)
- curl documentation
- LibriTTS-R paper (arXiv 2305.18802)
- FFmpeg documentation
- TRON developer hub: TRC-20
- W3C WAI: making audio and video accessible
- LibriVox public-domain audiobooks
- OpenAI API reference: create speech
