About AuricText: Independent TTS API for Startups

By Lvl2 editorial team · Updated 2026-11-19

AuricText is an independent text-to-speech API built for startup teams who need natural-sounding voice interfaces without enterprise complexity. We strip away the bureaucracy to help you prototype and deploy IVR systems quickly, focusing on the specific needs of small, agile teams rather than large corporate structures.

This page explains who we are, how our voices are licensed, and what technical limits to expect. You will find details on our data privacy policy, support channels, and independent status relative to major brands. Note that support is by email only. Read on to understand exactly what AuricText offers and how it fits into your development workflow.

About AuricText: Independent TTS API for Startups

Frequently asked questions

Who runs this website?

This is an independent storefront for the AuricText API. We are not affiliated with OpenAI, ElevenLabs, or other major brands. Our team builds and maintains this specific interface to provide text-to-speech services directly to startups and developers without enterprise complexity.

Are the voices copyrighted?

Our voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and public-domain audiobook readings from LibriVox. While the source material has specific licenses, the generated audio files are yours to use. We do not offer custom voice cloning for customers.

Do you store my text data?

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, characters, and status, solely for billing purposes.

Is there an SLA guarantee?

No. We do not offer an SLA, uptime guarantee, or enterprise contract. Requests are processed sequentially, so under load, a request may wait in a queue. Typical speed is roughly real time, but short phrases may take a few seconds.

How do I contact support?

Support is by email only. We do not have a phone number or live chat. You can reach us via the contact page or by emailing the support address provided in your account dashboard. We aim to respond to all inquiries promptly.

Are you connected to OpenAI?

No. We are not affiliated with OpenAI. Our API endpoint shape matches the OpenAI speech endpoint, so their SDKs work by changing the base URL. However, we are a completely independent service with our own pricing, voices, and infrastructure.

Can I use voices commercially?

Yes. The voices are based on LibriTTS-R (CC BY 4.0) and public-domain readings. You can use the generated audio for commercial purposes, including IVR systems and marketing. You can use the generated audio for commercial purposes, including IVR systems and marketing.

Do you offer refunds?

Failed requests are free; if synthesis fails, the charge is returned in full. We do not offer general refunds for successful generations. Your balance does not expire, and you can top up with USDT on the TRON network at any time.

Is the data encrypted?

We use standard API authentication via Bearer tokens. For specific encryption standards, please contact support via email.

Where are servers located?

The provided facts do not specify server locations. We do not offer on-premise installation or data-center choice. Our cloud API processes requests sequentially, and we do not guarantee specific geographic locations for the infrastructure hosting your text-to-speech generations.

Do you have a privacy policy?

Yes, we have a privacy policy available on our site. It details how we handle input text, request metadata for billing, and your account information. We do not sell user data, and input text is not used for training models.

Can I sue if audio fails?

We do not offer an SLA or enterprise contract. If a request fails, you are not charged. We do not provide legal guarantees or indemnities. Disputes are handled via email support. There is no on-premise option or certified compliance.

Do you sell user data?

No. Input text is used only to produce audio and is not used for training. We keep request metadata for billing purposes only. We do not sell user data to third parties. Your privacy is respected within the scope of our service.

Who owns the generated audio?

You own the generated audio files produced by your requests. The voices are based on licensed recordings, but the output is yours to use. We do not claim ownership over the specific audio files you generate, provided you have paid for the characters.

Who we are

Sound wave illustration for: Who we are

AuricText is an independent storefront for a text-to-speech API built for startup teams running IVR systems. We deliver natural, human-sounding text-to-speech audio that keeps callers engaged. Our focus is on startup needs, stripping away enterprise complexity so you can prototype and deploy voice interfaces fast.

We prioritize speed and simplicity. By removing corporate bureaucracy, we ensure our API remains straightforward and affordable for small teams.

Independent editorial team

Independent editorial team

Our team operates independently, allowing us to make quick decisions without waiting for multiple approval layers.

Focus on startup needs

Focus on startup needs

Our API is designed specifically for call centers and IVR systems where cost and integration speed matter. We provide clear documentation and a straightforward pricing model that scales with your usage, avoiding hidden fees or complex tier structures.

No corporate bureaucracy

Unlike large providers, we do not impose rigid enterprise contracts or minimum spend requirements. You can start with a free trial and scale as your business grows, paying only for the characters you actually use.

Voice licensing and origins

Sound wave illustration for: Voice licensing and origins

AuricText’s voice engine relies on licensed recordings sourced directly from the LibriTTS-R dataset. By utilizing these specific recordings, we ensure a consistent and high-quality synthetic voice experience for startup teams building IVR systems without the overhead of enterprise licensing complexities.

LibriTTS-R dataset usage

Our eight voice presets—Aria, Nova, Luna, Sage, Felix, Milo, Atlas, and Orion—are synthesized from audio samples within the LibriTTS-R collection. This dataset provides the foundational acoustic features that allow our API to generate natural, human-sounding speech across 23 languages, ensuring reliable performance for your call center applications.

  • Sourced from LibriTTS-R collection
  • Covers 8 distinct voice presets
  • Enables multi-language synthesis

CC BY 4.0 license details

The underlying audio data is distributed under the Creative Commons Attribution 4.0 International license. You can use the generated audio for commercial projects. The voices are based on licensed recordings, but the output is yours to use. We handle the licensing complexity so your team can focus on integration and deployment.

LibriVox public domain sources

Many of the base recordings originate from LibriVox, a platform dedicated to public domain audiobook readings. By leveraging these public domain sources, we maintain a transparent and legally sound foundation for our TTS engine, offering startups a clear path to scaling their voice interfaces without hidden royalty fees.

  • Roots in LibriVox public domain
  • Transparent legal foundation
  • No hidden royalty fees

Data privacy policy

Sound wave illustration for: Data privacy policy

AuricText treats your input data with strict minimalism. When you send text for synthesis, we use it solely to generate the requested audio file. Once the audio is delivered, the original text is not stored or retained on our servers. This ensures that your voiceovers, IVR scripts, or customer communications remain transient and private.

Text used only for generation

We process your input text exclusively for the purpose of creating the audio response. The text is not archived, indexed, or repurposed for any other function. This approach keeps your data lightweight and reduces the attack surface, which is critical for startup teams managing sensitive customer information or proprietary scripts.

  • Input text is processed in real-time for audio synthesis only.
  • No long-term storage of the source text after generation.

No training on input data

A common concern with AI voice services is the use of customer data to improve models. With AuricText, your input data is strictly excluded from training datasets. Whether you use the free trial or a paid USDT top-up, your texts do not contribute to the model's learning process, ensuring your intellectual property remains yours.

  • Input text is never used for model training.
  • Your data remains isolated from our training pipelines.

Metadata retention for billing

While the text itself is discarded, we retain request metadata including timestamp, character count, and status. This information is kept solely for accurate billing and usage tracking, allowing you to verify charges against your usage dashboard. This metadata is essential for our transparent, character-based pricing model without retaining the content of your messages.

  • Metadata (time, characters, status) is kept for billing.
  • Input text is not stored; only usage logs remain.

Honest technical limits

Sound wave illustration for: Honest technical limits

AuricText is built for straightforward integration, which means we prioritize reliable batch processing over complex real-time features. We strip away enterprise bloat to keep your IVR prototypes simple and fast, ensuring you get audio files without unnecessary protocol overhead.

No streaming support

We return complete audio files via the API response. There is no WebSocket or streaming endpoint to receive partial audio chunks as they are generated. This design keeps the implementation simple for IVR systems that expect a full file to play.

No SSML or emotion tags

Our API accepts plain text input. We do not support SSML markup, phoneme dictionaries, or custom pronunciation lexicons. There are no emotion tags to modulate tone; each voice preset has a fixed character that applies to the entire request.

Queue delays under load

Requests are processed one after another. While typical generation speed is roughly real-time (25 characters per second), short phrases may take a few seconds. Under heavy load, requests may wait in a queue before synthesis begins, so plan your timeouts accordingly.

Support and contact

Sound wave illustration for: Support and contact

We keep things simple so you can focus on building your IVR. Our support is straightforward and accessible via email for any questions about the API, billing, or voice selection.

Email only support channel

Send your queries to our support email. We handle all technical and account-related questions through this single channel to maintain clarity and efficiency.

No live chat or phone

There is no live chat widget or phone number to call. We avoid the overhead of real-time support queues, keeping our operational costs low and passing those savings on to you as a startup-friendly API provider.

Response time expectations

We aim to respond to emails within one business day. Since we are a small independent team, complex technical issues may take slightly longer, but we prioritize clear, actionable answers over quick, vague replies.

  • No 24/7 emergency line
Support and contact

Independent status

Sound wave illustration for: Independent status

AuricText operates as a completely independent entity, distinct from the major tech giants that dominate the voice market. We built this API specifically for startup teams who need a reliable, straightforward text-to-speech solution without the baggage of corporate ecosystems. Our infrastructure is designed to be lightweight and direct, ensuring that your integration remains simple and your costs predictable.

Not affiliated with OpenAI

While our API endpoint structure mimics the OpenAI POST /v1/audio/speech format for ease of use, we are not OpenAI. The tts-1 and tts-1-hd model names are our own presets, using the same engine and price as OpenAI's, but not tied to their infrastructure. You can use the official OpenAI SDK by simply pointing it to our base URL, but we are a separate company with our own pricing, limits, and voice models.

  • Compatible request shape, separate infrastructure
  • Own voice presets and pricing
  • No dependency on OpenAI accounts or keys

No relationship with ElevenLabs

We are not ElevenLabs. Our voices are derived from the LibriTTS-R dataset and public-domain LibriVox recordings, offering a distinct sound profile. Unlike ElevenLabs, we do not offer custom voice cloning or emotion tags. We focus on providing eight stable, licensed presets that are consistent and predictable for IVR and call center applications.

  • Distinct voice origin from LibriVox
  • No voice cloning or emotion tags
  • Stable, licensed presets only

Standalone API infrastructure

Our API is a cloud service. We do not rely on third-party cloud voice providers to generate audio. This means we control the queue, the rate limits, and the billing directly. You get a single API key that works across all our endpoints, with transparent usage tracking and USDT-based top-ups that don’t expire.

  • Self-hosted generation engine
  • Direct billing via USDT (TRC-20)
  • No third-party voice dependencies

Pricing and payment

Sound wave illustration for: Pricing and payment

Pricing is straightforward: $10 per 1 million characters, which equals $0.01 per 1,000 characters. Spaces and punctuation count toward this total. Failed requests are free, with charges returned in full if synthesis fails. Startups get a $0.50 trial credit (valid for 7 days) when signing in with Google.

There are no monthly plans, seats, or minimum spends. You only pay for what you generate. Balances do not expire, and you can top up as needed without waiting for a billing cycle.

$10 per million characters

Our rate is $10 per 1 million characters, or $0.01 per 1,000 characters. Spaces and punctuation count as characters. If a request fails, you are not charged. This predictable model helps startup teams budget their IVR and voice interface projects without hidden fees.

USDT (TRC-20) top-ups only

We accept USDT on the TRON network (TRC-20) for top-ups. The minimum deposit is $5. Credits appear automatically within minutes after your transfer. We do not process cards, PayPal, or issue invoices with VAT, keeping the process simple for global teams.

No monthly plans or subscriptions

We do not offer monthly plans, subscriptions, or seat limits. There is no minimum spend required to keep your account active. Your balance does not expire, allowing you to use credit exactly when your project demands it.

Technical specifications

Sound wave illustration for: Technical specifications

AuricText provides a robust API for generating natural-sounding speech. Our system supports 8 distinct voice presets across 23 languages, ensuring clear communication for your IVR systems. We accept MP3, WAV, OPUS, FLAC, and PCM formats for maximum compatibility.

8 voice presets available

Choose from four female voices (Aria, Nova, Luna, Sage) and four male voices (Felix, Milo, Atlas, Orion). Each voice offers a distinct tone, from warm narration to energetic promo styles.

  • 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

Our API supports 23 languages, allowing any voice to speak any supported language. If the language is not specified, it is guessed from the script, with Latin text defaulting to English.

  • Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, Turkish

MP3, WAV, OPUS, FLAC, PCM formats

Select the response format that best fits your infrastructure. We support MP3, WAV, OPUS, FLAC, and PCM, giving you flexibility in how you handle generated audio files.

Listen to the 8 voices

Each sample below was generated by this API from the same sentence. Its predictable pacing ensures that complex instructions are delivered clearly, reducing the chance of customer confusion.

Avatar of the voice aria
ariafemale
Avatar of the voice nova
novafemale
Avatar of the voice luna
lunafemale
Avatar of the voice sage
sagefemale
Avatar of the voice felix
felixmale
Avatar of the voice milo
milomale
Avatar of the voice atlas
atlasmale
Avatar of the voice orion
orionmale

23 languages

Any voice speaks any of these languages.

Sources and further reading

  1. Wikipedia
  2. curl documentation
  3. Python documentation
  4. LibriTTS-R corpus (OpenSLR 141)
  5. Node.js documentation
  6. ISO 639 language codes
  7. openai-node on GitHub
  8. Creative Commons Attribution 4.0
  9. Tether (USDT)
  10. LibriTTS-R paper (arXiv 2305.18802)
  11. IETF RFC 7845: Ogg Opus
  12. TRON developer hub: TRC-20

More on this site