Frequently asked questions
Who runs this site?
This is an independent storefront for the FitVoice TTS API. We are a standalone team providing a straightforward API to turn scripts into audio files. We operate separately from major tech brands and focus on delivering reliable synthetic voices for fitness and meditation applications without complex integrations.
Are you affiliated with OpenAI?
No. We are not affiliated with OpenAI, ElevenLabs, Google, Amazon, Microsoft, or any other brand mentioned in comparisons. Our API endpoint structure mimics the OpenAI speech endpoint for convenience, but we are an independent provider. We do not use OpenAI’s infrastructure or branding.
Is my data safe?
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 (time, characters, status) solely for billing purposes. We do not claim specific certifications like SOC 2 or HIPAA.
Do you sell my data?
No. 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 retain request metadata for billing. We do not sell your data or share it with third parties for marketing.
Can I request data deletion?
We keep request metadata for billing purposes. Contact us via email for specific inquiries regarding your account data retention.
How are voices licensed?
Voices are based on licensed recordings from LibriTTS-R (CC BY 4.0), sourced from LibriVox public-domain audiobook readings. Every generated file carries an inaudible watermark that marks it as synthetic speech. This ensures transparency about the synthetic nature of the audio produced by our API.
Is the site secure?
We use standard API authentication with Bearer tokens. We do not claim an SLA or uptime guarantee. Support is provided via email only. We do not have a physical address or phone number listed. For security specifics, refer to our general data handling practices.
Do you have a physical address?
No. We do not list a physical address. We are a remote team operating as an independent storefront. Support is available via email only. We do not have a phone number or live chat. All communication regarding your account or technical issues goes through email.
Who is the editor?
The editorial content is managed by the FitVoice product team. We are an independent group focused on providing reliable synthetic voices for fitness and meditation apps. We do not list individual editors or team members publicly. Our focus is on the API functionality and clear documentation.
Can I suggest a voice?
We offer eight fixed presets: Aria, Nova, Luna, Sage, Felix, Milo, Atlas, and Orion. We do not offer custom voice cloning or uploading your own voice. If you need a different voice, you must choose from the available presets or adjust speed and language settings.
Do you hire voice actors?
Our voices are based on existing licensed recordings from LibriVox public-domain audiobook readings. We do not currently hire new voice actors for custom recordings. The eight available presets are derived from these existing public-domain sources, ensuring consistent quality and licensing for our API users.
Is the team remote?
Yes. We are an independent team operating without a listed physical address. Support is provided via email only. We do not have a phone number or live chat. Our operations are streamlined to provide a straightforward API experience for product teams in fitness and meditation.
How do I contact support?
Support is available by email only. We do not have a phone number or live chat. You can reach us through the contact page on this site. We do not offer an SLA or uptime guarantee. Please include your API key and details in your email.
Who We Are
FitVoice is built by an independent editorial team focused on the fitness and meditation sectors. We provide a straightforward API to turn scripts into audio files without complex integrations. Our goal is to give product teams reliable, calm synthetic voices for app audio.
We operate as a standalone storefront for the same text-to-speech API, distinct from larger technology providers. Our team curates the technical details and pricing specifically for app developers who need consistent audio output.
Unlike many competitors, we do not offer streaming, SSML, or custom voice cloning. We stick to what works: simple HTTP requests, clear pricing, and no hidden enterprise contracts.
Independent editorial team
Our team manages the API infrastructure and documentation directly. We handle key issuance, billing via USDT on the TRON network (TRC-20), and voice selection without a large corporate overhead.
We prioritize clarity over hype. Every feature is listed, and every limitation is stated plainly. If a feature isn't listed, it doesn't exist.
- Direct API access
- No middlemen
- Transparent billing
Focus on fitness apps
We designed our voice presets and documentation for fitness and meditation applications. Voices like Luna (soft, storytelling) and Sage (calm, instructional) fit well with guided workouts.
Our pricing structure supports high-volume usage typical of mobile apps, with no monthly plans or minimum spend requirements.
We understand that app developers need predictable audio generation speeds and reliable endpoints.
- Optimized for mobile apps
- Calm voice presets
- Simple integration
No affiliation with big brands
We are not affiliated with OpenAI, Google, Amazon, Microsoft, or ElevenLabs. Our API endpoint and SDK compatibility are based on the OpenAI request shape.
We do not use their branding or claim to be their official provider. We simply use a compatible request shape for ease of integration.
Our voices are licensed from LibriTTS-R and LibriVox, separate from any major tech company's voice library.
- No OpenAI affiliation
- No Google affiliation
- No Amazon affiliation
Voice Licensing
FitVoice voices are derived from licensed recordings, ensuring high-quality synthetic speech for fitness and meditation applications without the need for complex voice design.
Our approach relies on established, accessible sources to maintain consistency and reliability across all generated audio files.
LibriTTS-R recordings
All eight voice presets, including Aria, Nova, and Atlas, are based on the LibriTTS-R dataset. This ensures that every synthetic voice maintains a natural, consistent tone suitable for guided workouts or calm meditation sessions.
CC BY 4.0 license
The underlying recordings are licensed under Creative Commons Attribution 4.0. This allows us to integrate these voices into our API while adhering to standard attribution practices.
LibriVox public domain
The source material originates from LibriVox, which provides public-domain audiobook readings. By leveraging these freely available recordings, we offer a robust TTS solution without the overhead of proprietary voice acquisition, keeping the API straightforward for your team.
Data Privacy
FitVoice treats your input text strictly as a production asset. We use your text solely to generate the audio file and never reuse it for model training or improvement. Input text is used only to produce the audio and is not used for training.
Text not used for training
When you send text to our API, it is processed to create the audio response. That text is not added to our training datasets, ensuring your unique voiceover scripts remain distinct from our model's future iterations.
Audio not stored
The generated audio file is returned directly in the API response and is not persisted on our servers after delivery. Generated audio is returned in the response and not stored.
Metadata kept for billing
We retain minimal request metadata, including the timestamp, character count, and status, solely for accurate billing and usage tracking. This data helps us calculate costs correctly and manage your account balance without storing the actual content of your requests.
API Limitations
FitVoice provides a straightforward API for fitness and meditation apps without the overhead of enterprise agreements. We prioritize simplicity and clear pricing over complex service guarantees, allowing product teams to integrate voice synthesis without negotiating custom terms.
No SLA guarantee
We do not offer a Service Level Agreement (SLA) that guarantees specific performance metrics or response times. Our infrastructure is designed for reliability, but we do not contractually promise a certain percentage of successful requests or latency thresholds.
No uptime promise
There is no official uptime guarantee for the API. While we aim for consistent availability, brief interruptions may occur during maintenance or unexpected issues.
No enterprise contract
We do not provide traditional enterprise contracts with custom volume discounts or dedicated account managers. All users operate under the same standard terms, paying only for the characters they generate. This approach keeps integration simple and costs transparent for startups and teams alike.

Contact Information
For technical questions or billing issues, our team is available via email. We prioritize clear, written communication to ensure every detail is documented for product teams managing their fitness app audio integrations.
We do not offer telephone support. Our workflow is designed around asynchronous resolution, allowing you to provide context and receive detailed answers without waiting on hold.
Email support only
Send your inquiries to the support address provided in your dashboard.
No phone number
There is no dedicated phone line for customer service. This keeps our operational costs low and ensures that support interactions are searchable and reproducible for your engineering team.
No live chat
We do not operate a real-time live chat widget on this site. For immediate status checks, refer to the /docs/ page; for complex issues, email is the most efficient channel.
Brand Independence
FitVoice is an independent storefront for a text-to-speech API, operating separately from major technology providers. We do not use the OpenAI, Google, or Amazon brands to describe our core service or infrastructure.
Our API endpoint and pricing structure are distinct from those of other voice synthesis companies. While our request format mimics the OpenAI speech endpoint for convenience, this is a technical compatibility choice, not an indication of ownership or partnership.
Not affiliated with OpenAI
Our API follows a similar JSON request shape to the OpenAI speech endpoint, allowing you to use their official SDKs by simply changing the base URL. This compatibility is purely technical; we are not owned by, partnered with, or endorsed by OpenAI.
We operate independently with our own keys, billing, and voice models. The similarity in endpoint structure is designed to lower integration friction for developers already familiar with OpenAI’s tools, but our service, pricing, and infrastructure are entirely separate.
- Base URL differs: https://api.bg4550.net/v1
- Independent key management and billing
- No endorsement or partnership with OpenAI
Not affiliated with Google
We offer Google sign-in for account creation, but this is only for authentication convenience. We do not use Google Cloud Text-to-Speech infrastructure, nor do we share data or revenue with Google.
Our voice models are based on LibriTTS-R recordings, not Google’s proprietary neural networks. We maintain full control over our voice presets and pricing, independent of Google’s ecosystem.
- Google sign-in is for authentication only
- Independent infrastructure and voice models
- No data sharing with Google
Not affiliated with Amazon
Our service is not part of Amazon Polly or any Amazon Web Services product. We do not use Amazon’s voice synthesis engines or infrastructure.
We operate as a standalone API provider, with our own server infrastructure and billing system. Any similarity in voice quality or features is coincidental, not the result of a partnership or technology license from Amazon.
- Independent server infrastructure
- Standalone billing system
- No technology license from Amazon
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. The voice handles factual statements with precision, avoiding emotional bias that might distract from performance metrics or health statistics.
23 languages
Any voice speaks any of these languages.
