Frequently asked questions
What is an automatic voice over generator?
CalmVoice is a cloud text-to-speech API that accepts text input and returns an audio file. It uses standard endpoints compatible with existing SDKs, allowing assistive device startups to integrate speech synthesis directly into their applications without managing complex infrastructure or hardware dependencies.
How does the API handle long texts for assistive devices?
The API accepts up to 4096 characters per request. For longer texts, your system should split the input into parts of this size, generate audio for each segment, and join the resulting files. This ensures consistent processing and avoids payload limits during synthesis.
Which voices are best for elderly users?
Aria offers a warm, clear narrator style, while Luna provides a soft, storytelling tone. Sage is calm and instructional, and Milo is friendly and conversational. These presets deliver consistent, compliant audio outputs suitable for assistive devices without requiring human intervention.
What languages are supported?
We support 23 languages: Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, and Turkish. Any voice can speak any of these languages, with optional language code specification.
Does the system require streaming?
No. The API returns complete audio files upon request completion. It does not support streaming, WebSocket connections, or real-time partial audio generation. Clients must wait for the full response before playing the synthesized speech in their assistive device applications.
How is pricing calculated per character?
Pricing is $10 per 1 million characters, or $0.01 per 1,000 characters. Spaces and punctuation count as characters. Failed requests are free, with charges returned in full if synthesis fails, ensuring you only pay for successful audio generation.
What happens if a request fails?
If synthesis fails, the charge is returned in full. You are not billed for failed requests. This ensures accurate billing for your assistive device applications, as you only pay for successfully generated audio files that meet the specified parameters.
Can I use the OpenAI SDK with this API?
Yes. The API uses the same request shape as the OpenAI speech endpoint. You can use official OpenAI SDKs by changing the base URL to https://api.mtown.info/v1 and using your CalmVoice API key for authentication in your integration.
Is there a free trial available?
Accounts created with Google sign-in receive $0.50 of credit (about 50,000 characters) valid for 7 days. Email/password accounts start without trial credit. Top-ups via USTRC-20 do not expire, allowing you to test the API before committing to paid usage.
How do I authenticate my requests?
Include an Authorization header with the value Bearer followed by your API key (e.g., Bearer tts-...). Each account has one active key, which can be regenerated at any time. New keys become active immediately, invalidating the previous one.
Are the voices synthetic or real human recordings?
Voices are based on licensed recordings from LibriTTS-R (CC BY 4.0), derived from LibriVox public-domain audiobook readings. Every generated file contains an inaudible watermark marking it as synthetic speech, ensuring clear distinction between AI-generated audio and original human recordings.
What audio formats are supported?
The API supports mp3, wav, opus, flac, and pcm formats. You can specify the desired response format in your request. These formats ensure compatibility with various assistive device hardware and software ecosystems used by elderly care applications.
How fast is the audio generation?
Generation is roughly real-time for long texts (approx. 25 characters per second). Short phrases take a few seconds. Requests are processed sequentially, so under load, requests may queue. This consistency supports reliable audio output for assistive devices.
Can I clone my own voice for the device?
No. CalmVoice does not offer custom voice cloning, voice uploading, or voice design. You must choose from the eight available presets. This ensures consistent, compliant audio outputs at scale without the variability associated with custom cloning processes.
First request
curl
curl https://api.mtown.info/v1/audio/speech \
-H "Authorization: Bearer $TTS_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"tts-1","voice":"aria","input":"Hello from the API.","response_format":"mp3"}' \
--output hello.mp3Core API Capabilities
CalmVoice provides a cloud-based text-to-speech API designed for assistive device infrastructure. The service accepts text input and returns synthesized audio files via a standard RESTful interface, enabling seamless integration into elderly care applications without complex middleware.
RESTful Endpoint Structure
Audio synthesis is triggered via a POST request to https://api.mtown.info/v1/audio/speech. The endpoint accepts a JSON body containing the text input, selected voice preset, and desired audio format. This structure mirrors standard OpenAI compatibility, allowing existing SDKs to function by adjusting the base URL to https://api.mtown.info/v1.
Character Limit Handling
Each request supports a maximum of 4,096 characters. For longer texts, such as extended medication instructions or daily news briefings, the system splits the input into manageable parts, synthesizes each segment individually, and joins the resulting audio files into a single continuous output.
Audio Format Options
The API supports multiple output formats to suit various device hardware configurations. Available formats include mp3, wav, opus, flac, and pcm. This flexibility ensures compatibility with both low-bandwidth wireless assistive devices and higher-fidelity stationary care units.
Voice Selection for Care Applications
Voice selection directly impacts user comprehension in assistive contexts. CalmVoice provides eight distinct presets. Each voice offers a consistent tonal quality, ensuring that audio outputs remain predictable across different device interactions.
The API allows precise matching of voice characteristics to specific application needs, from soothing background narration to clear, directive instructions. This flexibility ensures that synthetic speech remains comfortable for extended listening periods without causing listener fatigue.
Calm Narrative Voices
Aria, Luna, and Sage offer warm, clear, and soft tones ideal for storytelling, bedtime stories, or gentle reminders. These voices minimize abrupt tonal shifts, providing a consistent auditory environment suitable for users with hearing sensitivities or cognitive decline.
Luna offers a soft, storytelling tone, while Sage provides a calm, instructional voice.
- Aria: Warm, clear narrator for general content
- Luna: Soft, storytelling tone for companion devices
- Sage: Calm, instructional voice for step-by-step guidance
Clear Instructional Tone
Milo provides a friendly, conversational tone, and Sage provides a calm, instructional tone. Their articulation ensures clear delivery of medication instructions, device prompts, or safety warnings.
These voices maintain a steady pace.
Sage is particularly suited for instructional content, delivering information with a measured, easy-to-follow rhythm that supports comprehension in noisy environments common in home care settings.
- Milo: Friendly, conversational tone for general prompts
- Sage: Calm, instructional voice for clear directives
- Optimized for readability in noisy environments
Neutral News Delivery
Atlas provides a neutral, news-style delivery ideal for information dissemination, weather updates, or daily briefings. Its consistent tone avoids emotional bias, making it suitable for factual content delivery where clarity is prioritized over expression.
Orion offers a deep, documentary voice for longer-form content.
Both Atlas and Orion maintain a steady cadence.
- Atlas: Neutral, news-style delivery for factual content
- Orion: Deep, documentary tone for longer narratives
- Consistent cadence reduces cognitive processing effort
Language Support for Global Devices
CalmVoice supports 23 languages, covering primary Western markets and key Asian regions for assistive device deployment. The system ensures consistent audio output across diverse linguistic environments without requiring separate infrastructure.
Support spans major European languages alongside critical Asian dialects, enabling global scalability for elderly care applications where multilingual accessibility is a regulatory or user-experience requirement.
Primary Western Languages
The API covers 15 Western languages including English, German, French, Spanish, Italian, Portuguese, Dutch, Russian, Polish, Turkish, Danish, Norwegian, Swedish, Finnish, and Greek. Each voice preset can articulate any of these languages with native-level clarity.
This breadth allows assistive device manufacturers to deploy a single hardware SKU across multiple Western markets without localized voice model dependencies.
- Arabic, English, French, German, Italian, Portuguese, Russian, Spanish, Dutch, Greek, Polish, Turkish, Danish, Norwegian, Swedish, Finnish
Asian Language Support
Support includes Chinese, Japanese, Korean, Hindi, and Malay. These languages are fully integrated into the 8 voice presets, allowing consistent tone selection across regions.
No additional language packs are required; the same API endpoint handles all supported scripts.
- Chinese, Japanese, Korean, Hindi, Malay
Language Detection Logic
If the language field is omitted, the API guesses the language from the script. Latin script text defaults to English. For non-English Latin-script text, you must explicitly set the language parameter.
Supported ISO codes ensure precise rendering. Languages such as Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, and Bengali are not supported.
Technical Specifications
CalmVoice delivers audio via standard HTTP responses, supporting MP3, WAV, Opus, FLAC, and PCM formats to suit diverse assistive device requirements. Requests are processed sequentially, meaning under load a request may wait in a queue before synthesis begins. Failed requests incur no charge.
Response Formats
Select the output format that best matches your device’s audio pipeline. We support MP3, WAV, Opus, FLAC, and PCM, allowing you to balance file size and quality for elderly care applications with varying hardware constraints.
Speed Adjustment Parameters
The API accepts a speed parameter ranging from 0.25 to 4.0, allowing you to adjust the narration rate for users with specific hearing or cognitive processing needs. Typical generation speed is roughly real-time, producing approximately 25 characters of text per second of audio on long texts, though short phrases may take a few seconds to process.
Authentication Security
Secure your integration with a single active API key per account, passed via the Authorization: Bearer tts-... header. You can issue a new key at any time, which instantly invalidates the previous one, ensuring tight control over access. Requests are limited to 60 per minute per key, scaling from 10 per minute for new accounts before the first top-up.
Use Cases in Elderly Care
CalmVoice enables startups to build reliable audio interfaces for elderly care applications. The API generates consistent, compliant audio outputs without human intervention, ensuring clear delivery of critical information to users with varying cognitive abilities.
Each request processes text through a dedicated engine, producing speech that remains intelligible even at adjusted speeds. This reliability supports continuous operation in assistive devices where audio clarity directly impacts user independence and safety.
Medication Reminders
Assistive devices can trigger audio notifications for daily medication schedules. The API converts schedule data into clear, audible cues using preset voices like Sage or Luna. These reminders provide audible instructions.
Speed adjustment allows caregivers to customize playback duration. Slower speeds enhance comprehension for users with hearing impairments, while faster speeds reduce interruption during routine activities. Each reminder is generated on-demand, keeping device memory usage minimal.
- Automated schedule-to-audio conversion
- Adjustable playback speed (0.25–4.0x)
- Clear instructional tone via Sage voice
- Minimal device storage requirements
Emergency Alert Systems
Emergency protocols require immediate, unambiguous audio output. CalmVoice delivers high-priority alerts using Atlas for neutral news-style delivery or Orion for deep, authoritative tones. These voices convey urgency.
The API handles short, critical phrases efficiently, generating audio within seconds. Failed requests incur no charge, ensuring reliable billing for essential safety features. Audio files are returned instantly, allowing devices to play alerts without delay.
- Urgent tone via Atlas or Orion
- Low-latency generation for alerts
- Fault-tolerant billing structure
- No emotional distortion in delivery
Companion Audio Devices
Companion devices provide continuous audio interaction for elderly users. CalmVoice supports natural conversation flow by generating responses to simple prompts. Voices like Aria and Nova offer warm, clear narration suitable for storytelling or ambient conversation.
Input text is processed without storing the audio file, reducing storage overhead on edge devices. This approach allows devices to generate fresh content dynamically, maintaining a sense of freshness and personalization without accumulating large media files on the hardware.
- Warm narration via Aria or Nova
- Dynamic content generation
- No persistent audio storage required
- Supports conversational flow

Integration Workflow
CalmVoice integrates with existing infrastructure using standard HTTP clients or the official OpenAI SDK. Developers configure the base URL to point to our API endpoint, enabling immediate compatibility with established codebases without requiring custom libraries or proprietary drivers.
SDK Compatibility
The API mirrors the request structure of major industry standards, allowing teams to leverage familiar SDKs. This reduces integration time for startups building assistive devices, as no custom client libraries are required for basic text-to-speech synthesis operations.
Base URL Configuration
Set your API base URL to https://api.mtown.info/v1. This endpoint handles all speech synthesis requests and ensures consistent routing for your application's audio generation needs across different device types.
Request Payload Structure
Send a JSON body via POST to /v1/audio/speech. Include the model, input text, voice preset, and response format. The system accepts texts up to 4096 characters per request, processing them sequentially for reliable output.
- POST https://api.mtown.info/v1/audio/speech
- JSON body with model, input, voice
- Authorization: Bearer <your-api-key>
Reliability and Performance
CalmVoice delivers consistent audio synthesis with generation speeds averaging roughly 25 characters per second for long texts. Short phrases typically require a few seconds to process. This throughput supports real-time requirements for assistive devices without requiring complex streaming infrastructure.
Under load, requests are processed sequentially, which may introduce queue delays. The system guarantees that failed synthesis requests incur no charge, ensuring financial reliability alongside technical stability for critical care applications.
Generation Speed Metrics
Synthesis operates at approximately 25 characters of text per second of audio output. This ratio holds true for extended texts, ensuring predictable latency for medication reminders or companion audio. Shorter inputs may take slightly longer per character due to initialization overhead.
Queue Management
Requests are processed sequentially. During periods of high demand, additional requests may wait in a processing queue before synthesis begins. This linear processing ensures consistent audio quality but may increase total response time under heavy load.
- Sequential processing ensures consistent output quality.
- Latency increases proportionally with queue depth.
- No prioritization tiers available.
Error Handling Protocols
If synthesis fails for any reason, the request is marked as failed and no characters are deducted from your balance. The full amount is returned automatically, ensuring accurate billing even during infrastructure disruptions.
- Failed requests are free.
- Balance is refunded in full.
- No partial charges for failed audio.
Pricing Structure
CalmVoice uses a straightforward per-character billing model at $10 per 1 million characters. This rate applies uniformly across all models and voices, ensuring predictable costs for high-volume assistive device deployments. Spaces and punctuation are counted as characters, simplifying cost estimation for your engineering teams.
Per-Character Billing
The standard rate is $0.01 per 1,000 characters. Failed requests are free, with the charge returned in full if synthesis fails. There are no monthly plans, seats, or minimum spend requirements, allowing startups to scale usage directly with demand without fixed overheads.
- Rate: $10 per 1 million characters
- Failed requests are free
- No monthly plans or minimum spend
Trial Credit Allocation
New accounts created via Google sign-in receive $0.50 in trial credit, valid for 7 days. This provides approximately 50,000 characters for initial integration testing. Accounts established with email and password do not receive this trial credit, requiring an immediate top-up to begin generation.
Top-Up Methods
Balances are topped up exclusively via USDT on the TRON network (TRC-20). The minimum top-up amount is $5. Credits are credited automatically within minutes of the transfer confirmation. There are no subscriptions, invoices, or VAT charges associated with these transactions.
- Currency: USDT (TRC-20)
- Minimum top-up: $5
- Processing time: automatic within minutes
Data Privacy and Compliance
CalmVoice processes input text exclusively for the purpose of generating the requested audio output. We do not utilize your text data for model training or other commercial purposes, ensuring that your application's content remains private and dedicated solely to its immediate synthesis function.
Input Data Usage
Text sent via the API is processed in real-time to produce the audio file. Once the synthesis is complete, the input text is not retained for training or secondary analysis, preserving the integrity of your specific use-case data.
Audio Storage Policy
Generated audio files are returned directly in the API response and are not stored on our servers. This ensures that your audio outputs remain transient, reducing the risk of long-term data exposure for sensitive assistive device applications.
Metadata Retention
We retain request metadata, including timestamp, character count, and status, strictly for billing and operational verification purposes. This minimal data set allows us to reconcile charges accurately without storing the actual content of your applications.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. You pay per character processed, not per voice type, ensuring predictable expenses as your user base grows.
23 languages
Any voice speaks any of these languages.
Sources and further reading
- Wikipedia
- Node.js documentation
- IETF RFC 7845: Ogg Opus
- openai-python on GitHub
- W3C WAI: making audio and video accessible
- OpenAI API reference: create speech
- Python documentation
- MDN: Web Speech API
- LibriVox public-domain audiobooks
- TRON developer hub: TRC-20
- LibriTTS-R paper (arXiv 2305.18802)
- ISO 639 language codes
