Frequently asked questions
Who runs this site?
This site is run by an independent storefront for the VocalFlow text-to-speech API. It is not affiliated with OpenAI, ElevenLabs, or other major brands. We operate as a direct provider of the TTS engine, managing keys, billing, and support independently for our users.
Are the voices original?
Our eight voice presets are based on licensed recordings from LibriTTS-R (CC BY 4.0) and public-domain audiobook readings from LibriVox. While the performances are licensed, each generated file includes an inaudible watermark to clearly mark the audio as synthetic speech produced by our API.
Can I use the voices commercially?
Yes. The underlying recordings are licensed under CC BY 4.0 or are public domain. You can use the generated audio for VODs, intros, and emote reactions. The cost is $10 per 1 million characters, with no monthly plans or additional licensing fees beyond the API usage. Note that generated files carry an inaudible watermark marking them as synthetic speech.
Do you store my text?
Input text is used only to produce the audio and is not used for training. Generated audio is returned in the response and not stored. We keep request metadata, such as time, character count, and status, solely for billing purposes and account management.
How do I contact support?
Support is available by email only. We do not offer a phone number or live chat. Please send your inquiries directly to our support email address. We aim to respond to all messages as quickly as possible.
Is the data secure?
We do not claim specific security certifications like SOC 2 or HIPAA. Input text is processed to generate audio and is not retained for training. We use standard authentication via Bearer tokens. For detailed data handling, please refer to our full privacy policy page.
Do you offer refunds?
Failed requests are free; if synthesis fails, the charge is returned in full. We do not offer refunds for successful generations. There are no monthly plans or subscriptions, so you only pay for what you use. Balance does not expire once credited to your account.
Are you affiliated with OpenAI?
No. We are an independent storefront. Our API endpoint mimics the OpenAI speech endpoint shape for compatibility, so the official OpenAI SDKs work by changing the base_url. However, we are not affiliated with OpenAI, ElevenLabs, Google, Amazon, Microsoft, or any other brand mentioned in comparisons.
Do you have a physical address?
We do not list a physical address on this site. We are an independent digital storefront operating the VocalFlow API. All interactions, including billing and support, are handled online via email. We do not have a physical office or warehouse for customer visits.
Can I request a new voice?
We currently offer eight fixed presets: Aria, Nova, Luna, Sage, Felix, Milo, Atlas, and Orion. We do not offer custom voice cloning or voice design services. Users cannot upload their own voice or request new specific voices; you must choose from the existing available presets.
Do you use AI to generate voices?
Our voices are based on licensed recordings from LibriTTS-R and LibriVox. The API synthesizes speech using these sources. While the core voice performances are derived from licensed human recordings, our specific underlying technology is not detailed further.
Is the service reliable?
We do not provide an SLA or uptime guarantee. Requests are processed sequentially, so under load, a request may wait in a queue. Typical speed is roughly real-time for long texts. Short phrases may take a few seconds. We strive for consistent performance but cannot guarantee it.
Do you have a privacy policy?
Yes, we maintain a privacy policy that details how we handle your data. Input text is not used for training, and audio is not stored. Request metadata is kept for billing. Please visit the privacy policy page for complete information on data retention and processing.
How do I report a bug?
Please report bugs via email support. We do not have a public issue tracker or dedicated bug reporting form. Include your API key, the text used, and the error details in your email. Our team will review the issue and respond directly to your inbox.
Our Mission
VocalFlow exists to give VTubers and streamers a reliable, no-code way to generate clear speech for VODs, intros, and emote reactions. We built a simple API that works with standard tools, so you can focus on your stream without wrestling with complex integrations.
Simple Tools for Streamers
Our interface is designed for creators who want to generate audio quickly without writing code. You can use our web tool for instant reactions or the API for automated overlays. Everything is built to fit seamlessly into your streaming workflow.
Focus on Quality
We prioritize clarity and natural pacing for live content. Our voices are licensed from the LibriTTS-R dataset, ensuring high-quality audio that sounds professional yet fits the VTuber aesthetic. Each generation is optimized for immediate playback in OBS or similar software.
Transparent Pricing
Pricing is straightforward: $10 per 1 million characters, with spaces and punctuation counting toward the total. Failed requests are free, and there are no monthly plans or hidden fees. You can top up with USDT (TRC-20) for instant credit.
Voice Licensing
We source voice data from the LibriTTS-R dataset, which is built on LibriVox audiobook readings. These recordings come from public domain texts, ensuring that the underlying source material is free from complex copyright restrictions.
LibriTTS-R Dataset
Our voices are generated from licensed recordings within the LibriTTS-R dataset. This provides a consistent, high-quality foundation for our synthetic speech engine without relying on proprietary voice models.
CC BY 4.0 License
The dataset is available under the Creative Commons Attribution 4.0 International license. While general attribution to VocalFlow is appreciated, strict compliance points to the source creators. Crediting the original speaker or recordist from LibriVox is not strictly required for API usage, though it is encouraged.
Public Domain Sources
Since the underlying texts used for training are in the public domain, you can use the generated audio for VODs, intros, and emote reactions. The source material is clearly licensed, though we do not guarantee that the generated audio itself is entirely free of all third-party copyright claims.

Data Policy
When you send text to our API, it is processed solely to generate the requested audio file. We do not use your input text or the resulting speech to train or improve our underlying models. Your content remains yours, ensuring your stream lines and VODs stay private.
No Training Data
The text you submit is used exclusively for synthesis. Once the audio is generated and returned in the response, the input data is not stored for long-term training purposes. This ensures that your specific voice lines or emote reactions are not repurposed for model development without your explicit consent.
No Audio Storage
Generated audio files are returned directly in the API response and are not persisted on our servers after delivery. You retain full control over your audio assets, whether you save them for VODs, intros, or live overlays. We do not host or archive your speech files.
Billing Metadata
We retain request metadata, including timestamps, character counts, and status codes, strictly for billing accuracy. This data allows us to process your USDT top-ups and track your usage against the 60 requests per minute limit. No other personal data is stored beyond what is necessary for account management.
Technical Approach
Our API follows the standard OpenAI speech endpoint structure, allowing you to use official OpenAI SDKs by simply changing the base_url to https://api.playb77.com/v1. This ensures immediate compatibility with existing codebases without requiring custom wrappers or learning a new proprietary protocol.
OpenAI SDK Compatible
You can integrate our service using the official OpenAI SDKs. Just point the base_url to https://api.playb77.com/v1 and use your tts-... API key. This method works for both Python and JavaScript clients, letting you generate audio with familiar syntax.
- Uses standard
POST /v1/audio/speechendpoint - Compatible with official OpenAI SDKs
- Requires only
base_urlconfiguration change
Standard JSON API
Requests are sent as simple JSON bodies containing model, input, voice, and optional parameters like speed or language. The API returns the audio file directly in the response, making it easy to pipe audio into stream overlays or save it for VODs.
- JSON request body structure
- Supports mp3, wav, opus, flac, pcm formats
- Returns audio data directly in response
No Custom SDK
We do not provide a proprietary SDK, mobile SDK, or WordPress plugin. We rely on the robust, widely-supported OpenAI SDK ecosystem. This keeps the integration lightweight and ensures you are using standard HTTP libraries without dependency on a single vendor's custom code.

Support and Contact
We keep communication simple and direct. All support requests go through email, ensuring your issue is logged and reviewed without the noise of live chat queues.
Email Only
Use the contact form or send mail directly to our support address. This is the single channel for billing questions, technical issues, or account inquiries. We do not maintain phone lines or social media support tickets.
No Live Chat
We do not offer real-time live chat on this site. This keeps our overhead low and allows us to focus entirely on building a reliable TTS API rather than staffing a 24/7 chat team.
Response Time
We aim to reply as soon as possible after receiving your message. Since we are a small team, response times vary depending on the volume of requests. For urgent billing issues, please include your account ID.

Limitations
VocalFlow is designed as a straightforward API for generating audio files, not for real-time broadcasting. We prioritize stable, high-quality synthesis over complex interactive features, keeping the tool simple for VTuber workflows.
No Streaming
The API returns complete audio files (mp3, wav, opus, flac, pcm) after processing. There is no WebSocket or streaming endpoint for partial audio chunks. Long texts are split into segments and joined server-side.
No SSML
We do not support Speech Synthesis Markup Language (SSML) tags for precise control over pauses, emphasis, or pronunciation. Speed and voice are the only adjustable parameters via standard JSON fields.
No Voice Cloning
You cannot upload a sample to clone a custom voice. The eight presets (aria, nova, luna, sage, felix, milo, atlas, orion) are fixed. While based on licensed recordings, they are not personalized clones of specific individuals.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Create your own virtual persona with animated avatars and start streaming to your audience today.
23 languages
Any voice speaks any of these languages.
