Cloud Text-to-Speech API: 8 Voices, 23 Languages

By VoiceCompliance editorial team · Updated 2026-11-19

This guide details the technical and creative specifications for the VoiceCompliance text-to-speech API, designed for product teams managing high-volume voice-over workflows. It outlines the eight synthetic voices, 23 supported languages, and pricing structure to help you select the optimal configuration for your accessibility and corporate audio needs.

Readers will find concrete examples of voice personalities, language handling, and audio formats, alongside integration specifics for the OpenAI-compatible endpoint. The page also covers cost efficiency at $0.01 per 1,000 characters, licensing terms, and troubleshooting steps for common integration issues, ensuring a clear understanding of the API’s capabilities and limits.

Cloud Text-to-Speech API: 8 Voices, 23 Languages

Frequently asked questions

Which voice is best for education?

Q: Which voice is suitable for education? A: Sage offers a calm, instructional tone, while Luna provides a soft, storytelling style. Both are suitable for educational content, ensuring clarity and consistency for accessible materials. You can adjust the speed between 0.25 and 4.0 to match the pacing required for your specific audience. Note that no SSML or custom pronunciation lexicons are available for fine-tuning specific terms.

Is HD voice worth the cost?

The tts-1-hd model uses the same engine as tts-1. The price remains $0.01 per 1,000 characters for both models. Since the cost is identical, HD voice provides the same engine performance.

Can I mix voices in one video?

Yes, you can generate separate audio files for each voice and combine them in your video editor. Each request produces a distinct audio file. You can utilize different voices like Aria for narration and Felix for promotional segments within the same project, ensuring consistent audio quality across all segments.

How accurate is language detection?

The API guesses the language from the script if not specified. Latin script text defaults to English. For non-English Latin-script text, you should explicitly set the language parameter. This ensures accurate synthesis for the 23 supported languages, including Arabic, Chinese, and Swahili, preventing mispronunciation in multilingual content.

Do voices sound natural?

Voices are based on licensed recordings from LibriTTS-R and LibriVox public-domain audiobook readings. Each generated file carries an inaudible watermark marking it as synthetic speech. The voices provide clear, consistent audio suitable for accessibility standards, though they are not intended to mimic real human emotions or celebrity voices.

Is there a best format?

You can choose from mp3, wav, opus, flac, or pcm. MP3 is efficient for web delivery, while WAV and FLAC offer lossless quality for editing. PCM is raw data. Select the format that matches your video production pipeline. All formats are supported by the standard API endpoint for consistent delivery.

Can I use voices for ads?

Yes, Felix is described as energetic and suitable for promo content. All voices are licensable synthetic voices derived from public-domain or licensed recordings. You can use them for commercial advertisements without additional restrictions beyond the API usage terms. The audio files are returned directly in the response for immediate use.

How do I choose a language?

Specify the ISO code in the request body. If omitted, the system guesses the language, defaulting to English for Latin script. For precise control, especially with non-English Latin text, set the language parameter explicitly. The API supports 23 languages, ensuring accurate pronunciation for your target audience across different regions.

Is the speech clear in noise?

The API generates clean audio files without background noise. Clarity depends on the voice selection and speed settings. You can adjust the speed between 0.25 and 4.0 to optimize intelligibility. The output is a standard audio file, allowing you to add background music or effects in your post-production workflow.

Can I change speed later?

Speed is determined at generation time via the speed parameter (0.25–4.0). Once the audio file is generated, the speed is fixed within the file. You cannot change the speed dynamically via the API after generation. To alter speed, you must regenerate the audio with the desired speed parameter.

Do voices have accents?

Voices are designed to be neutral or match their described style, such as Aria being a warm narrator. They do not have specific regional accents unless implied by the voice design. The language parameter ensures correct pronunciation for the 23 supported languages, maintaining consistency across different linguistic contexts in your content.

Is there a limit to video length?

Each request accepts up to 4,096 characters. For longer videos, split the text into parts and join the audio files. There is no hard limit on total video length, only on the character count per request. This allows you to generate audio for videos of any duration by batching requests.

Can I download the audio?

Yes, the API returns the audio file directly in the response. You can save this file to your local storage or upload it to your video production pipeline. The audio is not stored by the service, so you must download and manage the files yourself after each successful request.

How do I test voices first?

Sign in with Google to receive a free trial of $0.50 credit, valid for 7 days. This provides approximately 50,000 characters for testing. You can use the API key to generate audio with any of the 8 voices and 23 languages to evaluate quality before committing to larger production volumes.

Understanding Voice Personalities

Sound wave illustration for: Understanding Voice Personalities

VoiceCompliance offers distinct voice personalities optimized for high-volume video voice-overs. Each preset delivers consistent audio output suitable for accessibility standards and dyslexia-friendly formats, ensuring clarity across diverse content types.

Warm Narrator Tone

Aria provides a warm, clear narrator voice ideal for educational content and instructional materials.

Bright Upbeat Style

Nova delivers a bright, upbeat style suitable for marketing videos and promotional content. The energetic delivery maintains audience engagement without requiring additional audio post-production.

Soft Storytelling Voice

Luna offers a soft, storytelling voice for narrative-driven content. This preset ensures smooth transitions between segments, supporting consistent pacing for corporate training materials.

Male Voice Options

Sound wave illustration for: Male Voice Options

The male voice presets in our text-to-speech API cover distinct professional tones suitable for high-volume voice-over workflows. Each voice is engineered for clarity and consistency, ensuring your video content maintains a uniform auditory identity across all languages.

Energetic Promo Voice

Felix delivers an energetic, promo-oriented tone ideal for marketing videos and dynamic product announcements.

  • Ideal for promotional spots
  • High energy delivery
  • Clear consonant articulation

Friendly Conversational

Milo offers a friendly, conversational style that works well for instructional content and user guides. His natural pacing aids comprehension, particularly for accessibility-focused materials where clear, approachable speech is prioritized over formal narration.

  • Natural conversational flow
  • Suitable for tutorials
  • Approachable tone

Neutral News Anchor

Atlas provides a neutral, news-like delivery suitable for corporate updates or factual reporting.

  • Neutral delivery style
  • Low listener fatigue
  • Professional corporate use

Female Voice Options

Sound wave illustration for: Female Voice Options

Female voices provide distinct tonal qualities for different production needs. Aria offers a warm, clear narrator style ideal for general content. Nova delivers a bright, upbeat tone suitable for energetic marketing materials.

Luna offers a soft, storytelling voice, while Sage provides a calm, instructional voice. These options ensure consistent audio character across high-volume video voice-overs, maintaining brand identity without requiring multiple talent bookings.

Calm Instructional

Sage delivers a calm, instructional tone optimized for clarity. This voice is effective for corporate training modules and educational content where precise articulation reduces cognitive load.

The steady pacing and neutral inflection help learners retain information. It works well for procedural explanations or compliance videos requiring a professional yet approachable delivery.

  • Optimized for instructional clarity
  • Consistent tone for training modules
  • Reduces listener fatigue

Deep Documentary

Orion, while classified as male, is often grouped in documentary contexts. However, Luna provides a soft, storytelling voice.

It is suitable for documentaries, case studies, or brand stories requiring a personal touch while maintaining professional standards.

  • Soft storytelling delivery
  • Emotional nuance for narratives
  • Engaging for case studies

Clear Standard

Aria offers a warm, clear narrator voice. This is the standard choice for general-purpose voice-overs, ensuring high intelligibility across diverse audiences. Her voice is versatile for news, updates, or general content.

Nova provides a bright, upbeat alternative for promotional content. All voices support all 23 languages.

  • High intelligibility
  • Versatile for general content
  • Supports 23 languages

Language Support

Sound wave illustration for: Language Support

VoiceCompliance supports 23 languages, enabling product teams to localize accessibility content and video voice-overs without managing multiple voice engines. The API accepts any of the 8 preset voices for any supported language, ensuring consistent audio quality across markets.

23 Supported Languages

Our API handles Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, and Turkish. Each voice is optimized for these languages, delivering clear synthetic speech suitable for dyslexia-friendly audio formats.

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

Automatic Language Detection

The API automatically detects the language from the input text script. For Latin-script text, it defaults to English. This simplifies integration for teams managing mixed-language content, reducing the need for manual configuration in every request.

Explicit Language Setting

For non-English Latin-script text, you must set the language parameter to the correct ISO code. Without this, the API may misinterpret the script and default to English, resulting in incorrect pronunciation. This ensures accuracy for languages like Dutch or German using Latin characters.

Audio Format Selection

Sound wave illustration for: Audio Format Selection

Select the response_format parameter to optimize audio delivery for your specific infrastructure. The API supports MP3, WAV, Opus, FLAC, and PCM, allowing product teams to balance file size, quality, and compatibility without additional conversion steps.

MP3 for Web

MP3 is the standard for web delivery, offering broad compatibility with all major browsers and media players. Use this format for general-purpose voice-overs where file size efficiency is prioritized over bit-perfect fidelity.

  • Universal browser support
  • Smallest file size for HTTP delivery
  • Ideal for standard web playback

WAV for Quality

WAV provides uncompressed, lossless audio, ensuring the highest possible fidelity for archival or professional editing workflows. This format is suitable when precise audio integrity is required for compliance or master copies.

  • Lossless compression
  • High fidelity for editing
  • No generation artifacts

Opus for Bandwidth

Opus delivers superior audio quality at significantly lower bitrates compared to MP3. This format is optimal for reducing bandwidth costs on high-volume deployments while maintaining clear, intelligible speech for accessibility applications.

  • High efficiency at low bitrates
  • Reduced bandwidth consumption
  • Optimal for mobile delivery

Speed and Pacing

Sound wave illustration for: Speed and Pacing

VoiceCompliance supports a speed multiplier ranging from 0.25x to 4.0x, allowing precise control over generation pacing. This parameter adjusts the duration of synthetic speech without altering voice characteristics, ensuring consistent audio quality across different playback rates.

Adjustable pacing is critical for accessibility and content efficiency. Teams can tailor output to match specific audience needs, from slow, deliberate narration for cognitive accessibility to rapid delivery for high-volume content reviews.

Adjustable Speed Range

The API accepts a speed field with values between 0.25 and 4.0. A value of 1.0 represents normal playback. Lower values stretch audio for clarity, while higher values compress it for speed. The engine maintains natural prosody across the entire range.

  • Minimum speed: 0.25x (quarter speed)
  • Maximum speed: 4.0x (four times normal)
  • Default speed: 1.0x

Slow for Clarity

Reducing speed to 0.5x–0.75x enhances intelligibility for listeners with processing differences or non-native speakers. This is standard for educational content and accessibility-compliant video voice-overs. The synthetic voice remains clear without excessive robotic artifacts.

  • Recommended for dyslexia-friendly audio
  • Ideal for instructional training modules
  • Ensures WCAG clarity standards

Fast for Reviews

Increasing speed to 2.0x–4.0x significantly reduces generation time and audio duration. This is efficient for internal review workflows, QA testing, or summarizing large text blocks. Note that request processing remains sequential, so speed affects output duration, not queue wait time.

  • Reduces final audio duration
  • Ideal for rapid QA cycles
  • Maintains voice consistency at high speeds

Technical Integration

Sound wave illustration for: Technical Integration

The API endpoint POST https://api.368g.info/v1/audio/speech accepts JSON bodies identical to the OpenAI speech endpoint. Official OpenAI SDKs function correctly by setting the base_url to https://api.368g.info/v1 and supplying your API key.

Authentication requires the Authorization: Bearer tts-... header. One active key per account ensures consistent billing and rate limiting. A new key invalidates the previous one immediately.

OpenAI-Compatible API

Integration requires only standard HTTP POST requests. The request shape matches the OpenAI specification, allowing developers to reuse existing client libraries without custom wrappers.

Parameter Configuration

Configure model ("tts-1" or "tts-1-hd"), input (text up to 4096 characters), voice (8 presets), and response_format (mp3, wav, opus, flac, pcm). The speed parameter accepts values between 0.25 and 4.0. The optional language field accepts an ISO code.

If language is omitted, the system guesses from the script. Latin script defaults to English; specify explicitly for other Latin-script languages.

  • Model: tts-1 or tts-1-hd
  • Input limit: 4096 characters
  • Speed: 0.25–4.0
  • Formats: mp3, wav, opus, flac, pcm

Error Management

Failed synthesis requests incur no charge. The API returns a full refund for errors, ensuring billing accuracy for high-volume production pipelines.

Rate limits enforce 60 requests per minute per key (10 before the first top-up). Exceeding this may queue requests sequentially.

Technical Integration

Accessibility Standards

Sound wave illustration for: Accessibility Standards

VoiceCompliance delivers consistent synthetic audio optimized for accessibility standards, ensuring clear output suitable for assistive technologies. Our API generates high-fidelity audio files that meet the clarity requirements expected in professional educational and corporate training environments.

WCAG Compliance

Generated audio supports WCAG guidelines by providing distinct, intelligible speech for screen readers and audio players. The consistent voice profiles ensure predictable listening experiences for users with cognitive or visual impairments.

Clear Audio Output

We prioritize intelligibility over stylistic flourish, producing audio that remains clear even at varying speeds. This clarity ensures that content remains accessible to dyslexic readers and non-native speakers listening to synthesized text.

Synthetic Watermark

Every generated file carries an inaudible watermark marking it as synthetic speech. This transparent marker helps compliance teams verify that audio content is AI-generated, supporting audit trails and regulatory transparency without affecting playback quality.

Cost and Pricing

Sound wave illustration for: Cost and Pricing

VoiceCompliance charges a flat rate of $0.01 per 1,000 characters, including spaces and punctuation. This predictable structure allows product teams to forecast API costs accurately without complex tiered pricing or hidden fees.

$0.01 per 1,000 Characters

Each request processes up to 4,096 characters. For longer texts, split the input and join the resulting audio files. The cost remains constant regardless of voice selection or audio format.

  • $0.01 per 1,000 characters
  • Spaces and punctuation count as characters
  • Flat rate for all models

No Hidden Fees

No monthly plans, seats, or minimum spend requirements exist. Balance does not expire. Top-ups via USDT (TRC-20) are credited automatically within minutes, with a $5 minimum deposit.

  • No monthly plans
  • No subscriptions
  • No VAT invoices

Free Failed Requests

If synthesis fails, the charge is returned in full. You only pay for successfully generated audio. This ensures you are not billed for processing errors or time-outs.

Use Case Matching

Sound wave illustration for: Use Case Matching

VoiceCompliance supports high-volume voice-over workflows for product teams, delivering consistent synthetic audio optimized for accessibility and dyslexia-friendly formats. Our API handles text up to 4,096 characters per request, ensuring reliable integration with your existing content management pipelines.

Educational Content

Generate clear, narration-style audio for e-learning modules using Luna or Sage voices. The API processes requests sequentially, making it suitable for batch-processing lesson scripts where consistency and clarity are prioritized over real-time streaming.

  • Use Luna for soft, storytelling narration in introductory modules.
  • Apply Sage for calm, instructional voice in technical tutorials.
  • Support 23 languages for global educational content distribution.

Corporate Training

Deliver uniform training materials across departments using Atlas for neutral news-style delivery or Milo for conversational instruction. With a rate limit of 60 requests per minute, teams can efficiently process bulk training scripts while maintaining predictable output costs.

  • Atlas provides a neutral tone ideal for policy updates.
  • Milo offers a friendly voice for onboarding sequences.
  • Adjust speed from 0.25 to 4.0 for varied pacing.

Marketing Videos

Create engaging promo content with Felix’s energetic style or Orion’s deep documentary tone. Each request generates a distinct audio file, allowing precise synchronization with video edits. Failed synthesis requests are free, reducing risk for high-volume campaign testing.

  • Felix delivers an energetic tone for promotional spots.
  • Orion provides a deep voice for documentary-style ads.
  • Output formats include MP3 for web and WAV for high quality.

Licensing and Rights

Sound wave illustration for: Licensing and Rights

VoiceCompliance delivers synthetic audio generated from licensed recordings and public-domain audiobook sources. The output audio files are royalty-free for commercial integration into your products, applications, or marketing materials without ongoing per-use fees.

Commercial Usage

You may embed the generated audio in paid products, video advertisements, or e-learning platforms without additional licensing costs. The API usage covers the right to distribute the synthesized speech as part of your commercial deliverables.

No Attribution Needed

Attribution to the original voice actors or the VoiceCompliance platform is not required for commercial deployments. You retain full control over how the audio is presented to your end-users.

Synthetic Watermark

Every generated audio file contains an inaudible watermark identifying it as synthetic speech. This feature supports transparency and compliance with emerging media authenticity standards without affecting playback or accessibility.

Troubleshooting Common Issues

Sound wave illustration for: Troubleshooting Common Issues

Product teams generating high-volume voice-overs must anticipate technical constraints to maintain consistent audio delivery pipelines. Misconfigurations in character limits or language settings often result in failed synthesis requests, delaying content production schedules.

Understanding the specific boundaries of the API ensures smoother integration with automated video editing workflows. Addressing these common issues proactively reduces operational friction and keeps voice-over projects on schedule.

Text Length Limits

Each request supports a maximum of 4,096 characters. If your script exceeds this limit, split the text into smaller parts and join the resulting audio files programmatically. This ensures consistent pacing without truncation.

  • Maximum 4,096 characters per request
  • Split long scripts into segments
  • Join audio files in your pipeline

Language Detection Errors

The API guesses the language from the script, defaulting to English for Latin text. For non-English Latin-script content, explicitly set the language parameter to avoid incorrect voice synthesis or errors.

  • Set language for non-English Latin text
  • Default is English for Latin scripts
  • Explicit settings prevent synthesis errors

Rate Limit Exceeded

Accounts are limited to 60 requests per minute per key. If you exceed this, requests queue or fail. Monitor your usage dashboard and adjust your generation speed or split tasks across time windows.

  • Limit: 60 requests per minute
  • Queue requests if limit is hit
  • Monitor usage via dashboard

Listen to the 8 voices

Each sample below was generated by this API from the same sentence. His natural pacing aids comprehension, particularly for accessibility-focused materials where clear, approachable speech is prioritized over formal narration.

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. openai-python on GitHub
  3. Tether (USDT)
  4. FFmpeg documentation
  5. LibriTTS-R paper (arXiv 2305.18802)
  6. IETF RFC 7845: Ogg Opus
  7. Python documentation
  8. LibriVox public-domain audiobooks
  9. Creative Commons Attribution 4.0
  10. LibriTTS-R corpus (OpenSLR 141)
  11. ISO 639 language codes
  12. openai-node on GitHub

More on this site