Text to Speech API for iPhone Apps and Call Center Menus

By LinguaLink editorial team · Updated 2026-11-19

This page provides the technical specifications and integration details for the LinguaLink text-to-speech API, designed specifically for indie developers building IVR systems and call center solutions. You will find the exact JSON endpoint structure, compatible voice profiles for call center menus, and the precise pricing model based on character consumption.

The content focuses on concrete implementation details, including supported audio formats, speed control ranges, and language support for English and European markets. It outlines the cloud-based deployment model, data privacy policies, and rate limiting rules, ensuring you have all the necessary information to integrate reliable voice synthesis into your bot or phone menu workflows without unnecessary complexity.

Text to Speech API for iPhone Apps and Call Center Menus

Frequently asked questions

What is the character limit per request?

Each API request supports a maximum of 4096 characters. For texts exceeding this limit, you must split the input into smaller parts and join the resulting audio files. Spaces and punctuation are counted toward the total character limit.

Which audio formats are supported?

The API supports mp3, wav, opus, flac, and pcm formats. You can specify the desired output format in the request body. These formats ensure compatibility with most IVR systems and call center software solutions.

How many voices are available?

There are eight voice presets available: aria, nova, luna, sage, felix, milo, atlas, and orion. These voices cover various tones suitable for different IVR contexts, from warm narrators to neutral news readers.

What is the speed range for synthesis?

The synthesis speed can be adjusted between 0.25 and 4.0. This allows you to control the playback rate of the generated audio to fit specific call center workflows or user experience requirements.

Does the API support streaming?

No, the API does not support streaming, WebSocket, or real-time partial audio. It returns the complete audio file after synthesis is finished. Requests are processed sequentially, so longer texts may experience queue delays.

Is SSML supported for formatting?

No, the API does not support SSML. It also lacks phoneme dictionaries, custom pronunciation lexicons, and emotion tags. Formatting is controlled via standard text input and speed parameters within the request body.

Can I clone my own voice?

No, custom voice cloning is not offered. You cannot upload your own voice or design custom voices. The eight available presets are based on licensed recordings from LibriTTS-R and LibriVox public-domain audiobook readings.

Are there monthly subscription plans?

No, there are no monthly plans, seats, or minimum spend requirements. You pay only for the characters you use. The balance does not expire, and you can top up as needed without subscription commitments.

How is payment processed?

Payments are processed via USDT on the TRON network (TRC-20). There are no cards, PayPal, or invoices. The minimum top-up is $5, and funds are credited automatically within minutes after the transfer is confirmed.

Is there an SLA guarantee?

No, there is no SLA, uptime guarantee, or enterprise contract. The API is designed for indie developers and does not offer on-premise installs, data-center choices, or certifications like SOC 2 or HIPAA.

What happens if a request fails?

If synthesis fails, the charge is returned in full. Failed requests are free, so you are not billed for unsuccessful attempts. This ensures you only pay for successfully generated audio files.

Can I use the OpenAI SDK?

Yes, the official OpenAI SDKs work by changing the base_url to https://api.bme688.info/v1. The request shape is the same as the OpenAI speech endpoint, allowing easy integration into existing developer workflows.

Are voices licensed or synthetic?

Voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain audiobook readings. Every generated file carries an inaudible watermark that marks it as synthetic speech, distinguishing it from human recordings.

Is there a free trial available?

Yes, signing in with Google provides $0.50 of credit, valid for 7 days. Accounts created with email and password start without trial credit. This allows you to test the API before making a top-up.

First request

curl

curl https://api.bme688.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.mp3

Infrastructure for Voice Output

Sound wave illustration for: Infrastructure for Voice Output

Text-to-speech output is delivered via a standard JSON API endpoint. Send a POST request to the speech endpoint with your text and receive an audio file directly in the response. This structure supports straightforward integration into existing backend systems without requiring complex middleware.

JSON API Endpoint

JSON API Endpoint

Requests are sent to https://api.bme688.info/v1/audio/speech with a JSON body containing the model, input text, voice selection, and desired audio format. The endpoint returns the audio file upon successful synthesis. Requests are processed sequentially, so under load a request may wait in a queue.

  • Endpoint: POST /v1/audio/speech
  • Input: JSON body with text and parameters
  • Output: Direct audio file return

OpenAI SDK Compatibility

The API mimics the OpenAI speech endpoint structure, allowing you to use official OpenAI SDKs by simply changing the base_url to https://api.bme688.info/v1 and updating the API key. This reduces development time for teams already familiar with OpenAI’s ecosystem, requiring minimal code adjustments to switch providers.

Batch Processing for IVR

For call center menus, longer texts are split into parts of up to 4096 characters and joined into a single audio file. This ensures consistent voice delivery across entire IVR flows without manual concatenation. Requests are processed sequentially, so under load, a request may wait in a queue before synthesis begins.

Voice Profiles for Enterprise Use

Sound wave illustration for: Voice Profiles for Enterprise Use

LinguaLink provides eight distinct voice presets. Each voice is derived from licensed recordings, ensuring a professional baseline for automated systems without the variability of real-time human performance.

Neutral News Voice

Atlas delivers a neutral, news-style delivery ideal for information-heavy IVR menus. Its consistent tone reduces listener fatigue during repetitive loop sequences, providing clear auditory cues for menu navigation without emotional distraction.

Friendly Conversational Tone

Milo offers a conversational style suited for bot responses and customer service interactions. The voice maintains approachability while remaining professional, making it suitable for handling inquiries in call center applications where tone consistency is critical.

Deep Documentary Style

Orion provides a deep, documentary-style voice appropriate for detailed explanations or premium service announcements. Its authoritative presence works well for delivering complex instructions or high-value information where clarity and gravitas are required.

Technical Specifications

Sound wave illustration for: Technical Specifications

Technical specifications define the operational boundaries of the LinguaLink TTS API. The system accepts standard HTTP requests and returns synthesized audio files directly in the response body, eliminating the need for secondary retrieval steps.

Supported Audio Formats

The API supports five standard audio formats: mp3, wav, opus, flac, and pcm. Developers can select the format that best balances bandwidth efficiency and compatibility with their specific telephony infrastructure or mobile client requirements.

Speed Control Range

Voice playback speed is adjustable via the speed parameter, accepting values between 0.25 and 4.0. This allows fine-tuning of call center menus or bot responses to match natural conversational pacing or accommodate specific latency constraints in IVR systems.

Character Limits

Each API request supports a maximum input of 4,096 characters. For longer texts, clients must split the input into segments, generate audio for each, and join the resulting files. Spaces and punctuation are counted toward this limit.

Language Support

Sound wave illustration for: Language Support

LinguaLink provides text-to-speech synthesis for 23 languages. The API supports English, German, French, Spanish, Japanese, Korean, and Chinese, among others.

This broad linguistic coverage allows indie developers to deploy multilingual call center bots with a single API integration. Whether handling inquiries in Arabic, Hindi, or Turkish, the system maintains the same reliability and character-based billing structure, simplifying infrastructure for international support teams.

English and European Languages

The platform supports 23 languages, including English, German, French, Spanish, Italian, Portuguese, Russian, and Polish. Each voice preset can articulate these languages natively, allowing for consistent tone across different regional markets. This eliminates the need for separate voice licensing per country in your IVR architecture.

  • Supports 19 European languages including English, German, French, Spanish, and Russian.
  • All 8 voice presets (e.g., Atlas for news, Felix for promos) work across these languages.
  • Consistent audio quality regardless of the target European market.

Asian Language Support

Asian language capabilities include Chinese, Japanese, Korean, Malay, Hindi, and others. These are essential for expanding call center reach into high-volume Asian markets. The API handles complex character sets and tonal variations reliably, ensuring clear comprehension for end-users without requiring custom phoneme dictionaries.

  • Supports Chinese, Japanese, Korean, Malay, and Hindi.
  • Native handling of non-Latin scripts without additional configuration.
  • Suitable for high-volume international call center operations.

Auto-Detection Logic

The API automatically detects language from the input script. For Latin-script text, it defaults to English unless the language parameter is explicitly set. For non-English Latin text, you must specify the language code to ensure correct pronunciation. Non-Latin scripts are detected automatically.

  • Auto-detects language from input script.
  • Defaults to English for Latin-script text.
  • Requires explicit language parameter for non-English Latin text.
  • Automatic handling of non-Latin scripts.

Deployment Model

Sound wave illustration for: Deployment Model

LinguaLink operates as a cloud-native API, delivering synthesized audio via standard HTTP responses. This architecture eliminates the need for local inference hardware, allowing indie developers to integrate voice output directly into call center bots and IVR systems without managing server infrastructure.

Our service scales automatically with request volume. There is no option for local installation; all processing occurs on our remote servers, ensuring consistent performance and reducing operational overhead for small engineering teams.

Cloud-Based Architecture

The API endpoint https://api.bme688.info/v1 handles all synthesis tasks. By offloading CPU-intensive audio generation to our cloud infrastructure, your application remains lightweight. This model supports high-concurrency IVR scenarios where multiple call center bots request audio simultaneously without local resource contention.

  • No local GPU or CPU requirements for synthesis
  • Automatic scaling during peak call center hours
  • Standard HTTPS connection for secure data transfer

No On-Premise Option

We do not offer on-premise installation or data-center choice. All text is processed on our remote servers, and generated audio is returned immediately in the response. This ensures that voice quality remains consistent regardless of your local network conditions.

  • No dedicated hardware licensing required
  • No local storage for audio files
  • Simplified maintenance with zero server updates

Instant Key Generation

Upon signing up via Google or email, you receive an active API key. There is no waiting period for approval. You can begin sending POST requests to /v1/audio/speech within seconds, integrating voice features into your iPhone apps or IVR menus without delay.

  • Immediate access after account creation
  • Single active key per account
  • Instant revocation if key is compromised
Deployment Model

Pricing Structure

Sound wave illustration for: Pricing Structure

LinguaLink charges a flat rate of $0.01 per 1,000 characters, with spaces and punctuation counted toward the total. Failed requests incur no charge, ensuring you pay only for successful synthesis.

Per Character Billing

The cost is calculated precisely at $10 per million characters. This linear model scales predictably with your call center volume, eliminating complex tier surprises.

  • Standard rate: $0.01 per 1,000 characters
  • Spaces and punctuation count as characters
  • Failed requests are free

Flat Rate Model

There are no monthly plans, seats, or minimum spend requirements. You retain full control over your budget without committing to recurring fees.

No Hidden Fees

Pricing is transparent with no invoices with VAT. Balance does not expire, allowing you to top up only when needed.

Developer Experience

Sound wave illustration for: Developer Experience

Integration relies on standard HTTP POST requests to the speech endpoint. Developers can utilize the official OpenAI SDKs by simply adjusting the base URL, or implement direct JSON payloads using any preferred language. The request structure remains identical to industry standards, ensuring minimal learning curve for existing teams.

Authentication is managed via a single active Bearer token. The API provides clear status codes for successful synthesis and specific error types for invalid inputs, allowing for straightforward debugging within automated call center workflows.

Standard HTTP Requests

Send text via JSON to the speech endpoint. The system accepts standard fields like model, input, and voice. No custom headers or complex authentication flows are required beyond the standard Bearer token.

Error Handling

The API returns distinct HTTP status codes for failures. If synthesis fails due to invalid parameters or server errors, the request is marked as failed and no characters are deducted from your balance.

  • 400 Bad Request: Invalid JSON or missing fields.
  • 401 Unauthorized: Expired or incorrect API key.
  • 500 Server Error: Temporary processing failure; no charge applied.

Rate Limiting

Each API key is limited to 60 requests per minute. New accounts start with a lower limit of 10 requests per minute until the first top-up is processed. This ensures stable performance for IVR systems.

  • Standard limit: 60 requests/minute.
  • Initial limit: 10 requests/minute.
  • Limit resets every 60 seconds.

Data Privacy

Sound wave illustration for: Data Privacy

We process input text solely to generate the requested audio file. The input is not retained for model training or used to improve the underlying voice engine after synthesis completes.

Input Usage Policy

Input text is used only to produce the audio and is not used for training. We do not store the raw text content, ensuring that your call center scripts or bot dialogue remain private.

Audio Storage

Generated audio files are delivered directly in the API response and are not stored on our servers. Each response is transient, meaning you retain full ownership and control over the resulting speech files without automatic archiving by LinguaLink.

Billing Metadata

We retain only the metadata necessary for accurate billing and rate limit enforcement. This includes the timestamp, character count, and request status. This data is used exclusively to calculate your usage charges and verify API key limits.

Use Cases

Sound wave illustration for: Use Cases

This API integrates directly into call center workflows, replacing static audio files with dynamic voice generation for phone menus and bot responses.

Developers can manage character limits and speed settings to ensure consistent audio quality across IVR systems and iPhone applications without additional infrastructure.

IVR Menus

Generate dynamic greetings and menu options in real-time. The API handles variable data injection, allowing callers to hear personalized information without pre-recorded assets.

  • Dynamic menu item announcements
  • Real-time data injection
  • Standard HTTP POST requests

Bot Responses

Convert text inputs from chat or voice bots into natural-sounding speech. Use the /v1/audio/speech endpoint to deliver immediate audio feedback to users.

  • Text-to-audio conversion
  • JSON API integration
  • Support for multiple languages

iPhone Voice Apps

Integrate TTS capabilities into iOS applications via standard HTTP requests. Developers can specify voice presets and audio formats to match app requirements.

  • iOS application integration
  • Configurable voice presets
  • Multiple audio formats

Listen to the 8 voices

Each sample below was generated by this API from the same sentence. For projects needing only basic IVR menus without custom cloning, this preset range is sufficient and reliable.

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

More on this site