Switching from Azure TTS to VoxMedica for Healthcare Audio

By VoxMedica editorial team · Updated 2026-11-19

This guide helps healthcare startups move from Azure TTS to VoxMedica’s straightforward API. We explain how our OpenAI-compatible endpoint simplifies integration for patient education materials, removing the complexity of enterprise contracts and tiered pricing.

You will find concrete comparisons of voices, rates, and migration steps. Learn how to update your code, handle character limits, and start using your free trial credit to test audio quality quickly without committing to long-term agreements.

Switching from Azure TTS to VoxMedica for Healthcare Audio

Frequently asked questions

Is the switch difficult?

Switching is straightforward because our API mimics the OpenAI endpoint structure. You simply update your base_url to https://api.9a9993.info/v1 and use the official OpenAI SDKs. The request shape remains identical, requiring only your new API key. This minimizes code changes for healthcare startups integrating patient education audio.

Do I lose my voices?

We offer eight specific presets like aria and atlas. However, these voices support 23 languages, including English, Spanish, and German. If your Azure voices were standard, you can likely find a close match among our licensed recordings from LibriTTS-R and LibriVox.

Is the audio quality similar?

Our voices are based on licensed recordings from LibriTTS-R and LibriVox, ensuring clear, natural speech. Each file carries an inaudible watermark marking it as synthetic. While we do not claim to be the fastest or most realistic, our output is designed for reliable patient education materials without enterprise complexity.

Can I keep my Azure key?

No, you must generate a new API key specific to VoxMedica. Each account has one active key, which you can issue anytime via the dashboard. Your Azure key remains valid for your existing Microsoft services, but you will use your new VoxMedica key for our API requests.

How long does migration take?

Migration typically takes minutes. Since our endpoint matches the OpenAI structure, you mainly update your base_url and authentication header. You can test with our free trial credit immediately. Most developers complete the switch during a single development session, allowing quick deployment of new audio features for patient guides.

Do I need to change code?

Minimal changes are required. You only need to update the base_url to https://api.9a9993.info/v1 and ensure you are using the correct API key. The request fields like model, input, and voice remain the same. Our service is compatible with official OpenAI SDKs, preserving your existing integration logic.

Is the pricing better?

Our pricing is $10 per 1 million characters, or $0.01 per 1,000 characters. Failed requests are free. We do not offer monthly plans or subscriptions, only pay-as-you-go with USDT top-ups. This transparent model helps healthcare startups control costs without enterprise contracts, paying only for successful audio generation.

What about language support?

We support 23 languages, including English, Spanish, German, French, and Chinese. Any voice can speak any supported language. If the language is not set, it guesses from the script, defaulting to English for Latin text. We do not support Vietnamese, Indonesian, or Thai, so verify your target languages before switching.

Can I use both APIs?

Yes, you can run Azure and VoxMedica simultaneously. Since we use standard HTTP requests, you can route different patient education modules to different providers. This allows gradual migration or A/B testing voices. Each API uses its own key, so there is no conflict between your existing Azure and new VoxMedica services.

Is there a migration guide?

We provide documentation at /docs/ with ready curl examples and SDK instructions. The endpoint structure mirrors OpenAI, making the transition intuitive. You can find specific details on request fields, supported formats, and authentication. No complex migration tools are needed; simply update your configuration to point to our API base URL.

Do I need a new account?

Yes, you need a separate VoxMedica account created at /get-api-key/. Accounts are per site, so your Azure credentials do not transfer. You can sign up with Google or email. Each account has its own balance, key, and usage history, ensuring clear separation between your existing and new audio services.

What if I make a mistake?

Failed requests are free, so errors in generation do not incur charges. If you send invalid text or parameters, you simply retry. You can also regenerate your API key instantly if needed. Our support is available by email for any technical issues during your integration or voice selection process.

Can I go back to Azure?

Yes, you can revert to Azure TTS at any time. Since we do not lock you into proprietary formats or long-term contracts, you can switch back easily. Your VoxMedica balance remains available if you return.

Is there a discount for switching?

We offer a free trial with $0.50 credit (about 50,000 characters) valid for 7 days when you sign in with Google. This allows you to test quality before committing. There are no specific switching discounts, but the transparent per-character pricing and no minimum spend make it easy to start small.

Why Switch?

Sound wave illustration for: Why Switch?

Healthcare startups often face complex billing structures with legacy TTS providers. Azure TTS charges vary significantly by voice tier and region, making budget forecasting difficult for patient education projects. VoxMedica offers a transparent, flat rate of $10 per million characters, simplifying cost estimation for scalable audio content.

Unlike enterprise-grade solutions requiring long-term commitments, our service is designed for flexibility. You pay only for what you use, with no minimum spend or monthly fees. This model suits startups that need reliable audio outputs without the overhead of enterprise contracts.

Simpler pricing

Azure TTS pricing is nuanced, depending on the voice type (Standard vs. Neural) and language. VoxMedica applies a uniform rate of $0.01 per 1,000 characters across all models. This consistency helps you predict costs for large volumes of patient guides or appointment reminders without tracking tiered usage.

  • Flat rate: $10 per million characters
  • No hidden fees for regional differences
  • Failed requests are free

No enterprise contracts

Traditional enterprise TTS solutions often lock you into annual agreements or require minimum commit levels. VoxMedica operates on a pay-as-you-go basis. You can start with a free trial credit and top up only when needed, giving you full control over your cash flow.

  • No monthly subscriptions
  • No seat limits
  • Balance never expires

OpenAI compatibility

If your team is already familiar with the OpenAI speech endpoint, migration is straightforward. VoxMedica uses the same request shape, allowing you to simply update the base_url in your existing SDKs. This reduces development time and leverages your team's existing expertise with standard HTTP libraries.

  • Same JSON body structure as OpenAI
  • Works with official OpenAI SDKs
  • Easy base_url swap

Endpoint Comparison

Sound wave illustration for: Endpoint Comparison

Azure TTS relies on a complex URL structure with specific regions and endpoint versions, requiring careful configuration for each deployment. This often involves managing multiple base URLs and understanding regional data residency constraints.

VoxMedica simplifies this by mimicking the familiar OpenAI endpoint structure. Our API base is https://api.9a9993.info/v1, allowing you to send requests to POST /v1/audio/speech with a standard JSON body.

Azure URL structure

Azure typically requires constructing URLs like https://{region}.api.cognitive.microsoft.com/sts/v1.0/issueToken or specific speech services endpoints. This fragmentation can complicate integration, especially when scaling across different geographic regions or managing multiple subscriptions.

  • Region-specific base URLs
  • Separate token endpoints
  • Complex query parameters

Our OpenAI-like URL

We consolidate access into a single, consistent endpoint: POST https://api.9a9993.info/v1/audio/speech. This aligns with the request shape used by OpenAI, making it intuitive for developers already familiar with modern TTS APIs. You simply change the base URL in your existing client configuration.

  • Single base URL
  • Standard JSON body
  • No region suffixes

SDK compatibility

Because our request shape matches the OpenAI speech endpoint, the official OpenAI SDKs work seamlessly. You only need to update the base_url to point to https://api.9a9993.info/v1 and use your VoxMedica API key. This reduces boilerplate code and accelerates integration for healthcare startups.

  • OpenAI SDK compatible
  • Simple base_url change
  • No custom client needed

Voice Differences

Sound wave illustration for: Voice Differences

Azure offers dozens of neural voices, but healthcare workflows often require a consistent, neutral tone for patient education. Managing multiple voice styles across different modules adds complexity to your integration stack.

Our 8 presets provide clear, distinct options without overwhelming choice. Each voice is licensed from LibriTTS-R recordings, ensuring high-quality, natural-sounding speech suitable for medical guides and instructions.

Azure voice variety

Azure provides a wide array of neural voices with varying accents and styles. While this variety is powerful, it requires careful selection to maintain brand consistency across patient-facing materials.

Our 8 presets

We offer eight specific voices. They range from the warm, clear narration of Aria to the calm, instructional tone of Sage. This curated selection simplifies decision-making for healthcare startups.

  • Aria: Warm, clear narrator
  • Sage: Calm, instructional
  • Felix: Energetic, promo
  • Atlas: Neutral, news

Style differences

Unlike Azure’s extensive catalog, our voices are designed for clarity and ease of use. They are based on licensed recordings from LibriTTS-R and LibriVox, ensuring legal compliance and consistent quality.

Pricing Comparison

Sound wave illustration for: Pricing Comparison

Switching from Azure TTS often reveals hidden costs in per-character billing. Azure charges vary by voice tier and region, creating unpredictable expenses for healthcare startups generating patient guides. VoxMedica simplifies this with a transparent, flat rate of $10 per 1 million characters, ensuring your budget matches your usage exactly.

Azure per-character cost

Azure pricing fluctuates based on the specific neural voice selected and the data center region. This complexity can make long-term cost estimation difficult for projects with variable text lengths, such as dynamic patient education materials.

Our flat rate

We charge $10 per 1 million characters, regardless of the voice model (tts-1 or tts-1-hd) used. Spaces and punctuation count as characters. Failed requests are refunded in full, so you only pay for successful audio generation.

  • No regional price differences
  • No tier-based markup
  • Refunds for failed synthesis

Free trial advantage

New accounts signed in with Google receive $0.50 in credit, covering approximately 50,000 characters. This allows you to test audio quality for patient guides without immediate payment. Email-based accounts start with zero credit but can top up easily.

  • $0.50 credit for Google sign-ins
  • Valid for 7 days
  • No credit card required

Migration Steps

Sound wave illustration for: Migration Steps

Migrating your healthcare TTS integration to VoxMedica is a straightforward process that minimizes downtime for your patient education platforms. By following these three steps, you can transition from Azure’s complex enterprise setup to our simpler, OpenAI-compatible endpoint structure.

This guide focuses on the technical changes required to ensure your audio generation pipeline remains functional and cost-effective for healthcare startups.

Get API key

Navigate to the /get-api-key/ page on 9a9993.info to create your account. You can sign up using Google for immediate access or email and password. Upon registration, you will receive a unique tts-... key and $0.50 in trial credit valid for 7 days.

This key is your single point of authentication. You can generate a new key at any time, which instantly invalidates the old one, ensuring security without needing to update multiple systems.

  • Sign up via Google or email on /get-api-key/
  • Receive your tts-... API key and $0.50 trial credit
  • No enterprise contract or minimum spend required

Update base_url

Update your SDK configuration to point to our OpenAI-compatible endpoint. Change your base_url to https://api.9a9993.info/v1. This allows you to use the official OpenAI SDKs without major code rewrites.

Ensure your authorization header uses Bearer <your_key>. The request shape remains identical to the OpenAI speech endpoint, supporting model, input, voice, and response_format fields.

  • Set base_url to https://api.9a9993.info/v1
  • Use Bearer <your_key> in the Authorization header
  • Maintain existing OpenAI SDK request structure

Test requests

Send a test request with a short patient education phrase. Verify that the audio is generated correctly and the character count is accurate. Remember that spaces and punctuation count toward the 4096-character limit.

Check the response format matches your requirements (mp3, wav, opus, flac, or pcm). If the synthesis fails, no charge is applied, allowing you to debug safely.

  • Send a test POST request with sample text
  • Verify audio format and character counting
  • Confirm no charge for failed requests

Code Example

Sound wave illustration for: Code Example

Switching to VoxMedica requires minimal code changes because we mimic the OpenAI endpoint structure. Your existing logic for handling audio responses remains largely intact, reducing migration friction for healthcare startups. This compatibility ensures that patient education audio generation continues without disrupting your current development workflow.

You can update your base URL and API key to start generating speech immediately. The following examples demonstrate how to adjust your Python and Node.js implementations to use our API for creating clear, synthetic audio files for medical guides.

Python SDK change

Update your OpenAI client configuration to point to our API base URL. Replace your existing API key with your VoxMedica key. The rest of the parameters like model, input, and voice work identically to the standard OpenAI TTS endpoint.

This allows you to keep your existing error handling and audio processing logic while benefiting from our simpler pricing and straightforward integration for patient education materials.

  • Set base_url to https://api.9a9993.info/v1
  • Use tts-1 or tts-1-hd models
  • Keep voice and input parameters unchanged

Curl example

You can test the API directly using a simple cURL command. This method is useful for quick verification of voice quality and format support before integrating into your main application.

Replace YOUR_API_KEY with your actual key. The response will return the audio file directly, which you can save or stream to your patient education platform.

  • Use POST https://api.9a9993.info/v1/audio/speech
  • Set Authorization: Bearer YOUR_API_KEY
  • Specify model, voice, and input in the JSON body

Node.js update

In Node.js, update your HTTP client or OpenAI SDK configuration to use our base URL. This ensures that all TTS requests are routed correctly to our infrastructure.

The response handling remains the same: you receive the audio data, which you can then save as an MP3, WAV, or other supported format for your healthcare application.

  • Update baseURL in your client config
  • Maintain existing request payload structure
  • Handle audio response data as before

Testing Strategy

Sound wave illustration for: Testing Strategy

Testing your integration ensures patient education materials are clear and compliant before full deployment. With a $0.50 free trial credit, you can generate approximately 50,000 characters of audio without financial risk. This allows you to validate voice selection and latency.

Focus on verifying that the synthetic watermark is present and that audio formats match your EHR system’s requirements. Test edge cases like mixed-language inputs or long consent forms to ensure the 4096-character limit is handled correctly via splitting.

Use free trial

Sign in via Google to activate your $0.50 trial credit. Use this balance to generate sample audio for common patient scenarios, such as medication instructions or post-op care guides. Since the trial lasts 7 days, you have ample time to iterate on voice settings without committing funds.

Remember that email-only accounts start without trial credit. If you registered with email and password, you may need a small top-up to begin testing, though the $5 minimum is low enough for initial validation.

  • Activate trial via Google sign-in
  • Generate ~50,000 characters of test audio
  • Validate latency for short clinical queries

Check audio quality

Listen for clarity in medical terminology. While our voices are based on licensed recordings, ensure the selected preset matches your intended tone—e.g., 'sage' for instructional content or 'aria' for warm narration. Verify that the inaudible watermark is present.

Test speed settings between 0.25 and 4.0. Slower speeds may improve comprehension for elderly patients, while faster speeds suit quick reference guides. Ensure the audio does not clip or distort at higher speeds.

  • Select voice preset matching tone
  • Adjust speed for patient demographics
  • Verify watermark presence

Verify formats

Confirm that the generated audio files are compatible with your delivery platform. We support mp3, wav, opus, flac, and pcm. MP3 is generally sufficient for web delivery, while WAV may be required for specific EHR integrations.

Test the response_format parameter in your API calls. Ensure that your application correctly handles the binary response and saves the file with the correct extension. Verify that file sizes are reasonable for your bandwidth constraints.

  • Choose format compatible with EHR
  • Test mp3 vs. wav outputs
  • Validate binary response handling
Testing Strategy

Handling Limits

Sound wave illustration for: Handling Limits

Understanding rate limits and queue behavior ensures your patient education audio generation remains reliable during peak usage. Knowing how to manage these constraints helps maintain a smooth workflow for healthcare startups integrating TTS without unexpected delays.

Rate limits

Your API key allows 60 requests per minute. Before your first top-up, this limit is stricter at 10 requests per minute. This ensures fair usage while you are testing with your free trial credit.

Queue management

Requests are processed sequentially. Under heavy load, your synthesis jobs may wait in a queue before generating audio. Plan your integration to handle these slight delays gracefully, especially when generating multiple patient guides simultaneously.

Splitting long texts

Each request supports up to 4096 characters. For longer medical documents, split the text into smaller parts. Generate each segment individually and join the resulting audio files to create a complete patient guide.

Support Options

Sound wave illustration for: Support Options

We offer straightforward support tailored for healthcare startups that need clear answers without enterprise complexity. Our documentation covers the API structure, voice presets, and migration steps, ensuring you can integrate VoxMedica smoothly into your patient education workflows.

For direct assistance, our support team is available via email. We prioritize specific, technical answers to help you resolve integration issues quickly, keeping your development cycle moving without unnecessary delays or generic responses.

Email support

Our primary support channel is email, ensuring you get thoughtful, detailed responses to your technical questions. Our team reviews your issue and provides a precise solution tailored to your API usage.

We focus on clear, actionable advice rather than generic troubleshooting steps. Whether you are adjusting speed parameters or handling character limits, our email support helps you resolve issues efficiently without the noise of broader customer service departments.

  • Direct email access for technical queries
  • Responses tailored to your specific API usage
  • No waiting for live agent availability

Documentation

Our documentation provides a complete reference for the API endpoints, voice options, and pricing structures. It includes ready-to-use code examples in Python and cURL, making it easy for developers to test the integration immediately.

The docs also cover migration tips, explaining how to adjust your base URL and handle rate limits effectively. This resource is designed to be a reliable companion throughout your development process, reducing the need for constant support inquiries.

No live chat

Unlike many enterprise platforms, we do not offer live chat or phone support. This approach allows us to maintain focused, asynchronous communication where you receive detailed, written responses that you can reference later.

This structure suits healthcare startups who prefer clear, documented solutions over real-time conversations. You can access our email support anytime, ensuring you get the assistance you need without the pressure of immediate responses during busy development periods.

Common Issues

Sound wave illustration for: Common Issues

Switching from Azure TTS often reveals subtle configuration gaps. Healthcare audio requires precise language handling and strict character boundaries to ensure patient education materials render correctly without truncation or formatting errors.

Understanding these common pitfalls helps you avoid silent failures during deployment. By addressing language detection, character limits, and format compatibility early, you ensure reliable audio generation for your medical guides.

Language detection

Azure often infers language automatically, but our API requires explicit language codes for non-English Latin text. If you migrate patient guides from English to Spanish or French without setting this field, the system defaults to English, causing pronunciation errors.

Set the language parameter explicitly for every request to ensure accurate accent and intonation for your target audience.

  • Default Latin script text generates English audio.
  • Explicit language code ensures correct pronunciation.
  • Check supported languages: Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, Turkish.

Character limits

Each API request accepts up to 4096 characters, including spaces and punctuation. Splitting your patient guides into chunks is necessary. You must join the resulting audio files client-side.

Longer texts generate at roughly real-time speed (25 characters per second). Short phrases may take a few seconds to process.

  • Max 4096 characters per request.
  • Spaces and punctuation count toward the limit.
  • Split and join audio files manually.

Audio format issues

We support mp3, wav, opus, flac, and pcm. Ensure your playback system handles these formats correctly. Unlike Azure, we do not offer streaming or WebSocket endpoints, so you must wait for the full file.

Failed requests are free, but successful ones are charged. Verify your format selection matches your frontend capabilities.

  • Supported formats: mp3, wav, opus, flac, pcm.
  • No streaming or WebSocket support.
  • Check frontend compatibility before deployment.

Final Checklist

Sound wave illustration for: Final Checklist

Before launching your healthcare audio integration, ensure your API key is active and correctly configured in your application’s environment variables. Verify that your balance covers the initial batch of patient education materials to avoid unexpected interruptions during critical delivery windows.

Confirm that your codebase points to the correct base URL and handles the JSON response format. A final review of these settings prevents deployment delays and ensures seamless audio generation for your users.

Key active ['Verify your API key is active in the dashboard. Generate a new key if the old one has expired or been revoked.', 'Ensure the key is stored securely in your environment variables, not hardcoded in your source code.', 'Test a simple request to confirm the key authenticates successfully against the endpoint.'] ['Check key status in /get-api-key/', 'Rotate keys if compromised', 'Store in environment variables']

Balance sufficient ['Confirm your USDT balance is sufficient for your projected usage. Remember, the free trial credit expires in 7 days.', 'Top up via TRC-20 if needed, as credits are credited automatically within minutes.', 'Monitor usage to ensure you do not run out of characters mid-campaign.'] ['Top up minimum $5 USDT', 'Check balance regularly', 'No expiration on balance']

Code updated ['Update your base URL to https://api.9a9993.info/v1 if using the OpenAI SDK.', 'Verify that your request payload includes the correct model and voice parameters.', 'Test a sample request to ensure audio generation works as expected before full deployment.'] ['Update base URL', 'Verify payload structure', 'Run test request']

Next Steps

Sound wave illustration for: Next Steps

Before sending your first patient audio, ensure your USDT balance is active. Transfer at least $5 to your TRC-20 wallet address to cover initial generation costs.

Once funded, update your application to point to the production endpoint. You can now generate clear, compliant audio files for your healthcare materials.

Top-up USDT

Send USDT (TRC-20) to your account wallet. The minimum top-up is $5. Credits appear automatically within minutes after the blockchain confirms the transaction.

  • Network: TRON (TRC-20) only
  • Minimum amount: $5 USD
  • No invoices or VAT

Go live

Switch your SDK base_url to https://api.9a9993.info/v1. Use your active API key for authentication. Your first request will generate audio instantly.

  • Update base_url in your code
  • Verify your API key
  • Send a test speech request

Monitor usage

Check your dashboard for character counts and balance. Requests are processed sequentially, so long texts may queue during high load.

  • Track characters consumed
  • Watch for queue delays
  • Review billing logs

Listen to the 8 voices

Each sample below was generated by this API from the same sentence. This consistency helps you predict costs for large volumes of patient guides or appointment reminders without tracking tiered usage.

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. LibriTTS-R corpus (OpenSLR 141)
  3. ISO 639 language codes
  4. Python documentation
  5. openai-python on GitHub
  6. TRON developer hub: TRC-20
  7. openai-node on GitHub
  8. W3C WAI: making audio and video accessible
  9. Creative Commons Attribution 4.0
  10. OpenAI API reference: create speech
  11. LibriVox public-domain audiobooks
  12. MDN: Web Speech API

More on this site