Text-to-Speech API for Animation: 8 Voices, 23 Languages

By Vocaltoons editorial team · Updated 2026-11-18

This guide helps animation studios and cartoon creators select consistent character voices using the Vocaltoons API. You will find practical steps for matching eight distinct voice presets to your cast, managing 23 languages, and integrating audio directly into your storyboard workflow.

Readers get concrete examples of voice selection, technical settings, and cost analysis tailored to prototype production. This page is designed for teams needing reliable, license-free TTS without expensive custom licensing, ensuring your characters sound distinct from the first render.

Text-to-Speech API for Animation: 8 Voices, 23 Languages

Frequently asked questions

Which voice is best for a child character?

We do not have a specific child voice preset. You can use Luna for a soft, storytelling tone or Aria for a clear narrator style. Adjust the speed parameter between 0.25 and 4.0 to alter pacing. Since voices are based on licensed recordings, you cannot clone a child’s voice. Use the /v1/voices endpoint to review all eight available presets for your storyboard needs.

What voice works best for a villain?

Orion offers a deep, documentary-style voice suitable for menacing roles. Atlas provides a neutral, news-like tone that can feel authoritative or cold. Felix is energetic and promo-focused, which might work for a quirky villain. You can adjust the speed to make the delivery feel more aggressive or slow. There are no specific emotion tags, so pacing defines the character.

Can I use the same voice for multiple characters?

Yes, you can use any of the eight voices for multiple characters. The API does not distinguish between character identities, only the voice preset. You can assign Aria to one character and Luna to another, or reuse Orion across different scenes. This allows consistent audio production without needing unique voice licenses for each minor role in your animation.

How do I choose a voice for a documentary?

Orion is designed with a deep, documentary tone. Atlas offers a neutral, news-style delivery suitable for factual content. You can adjust the speed between 0.25 and 4.0 to control the pacing. Use the /v1/voices endpoint to listen to samples. These voices are based on licensed recordings, ensuring clear audio for your production without additional licensing fees.

What is the best voice for narration?

Aria is described as a warm, clear narrator. Luna offers a soft, storytelling tone ideal for narrative arcs. Sage provides a calm, instructional voice suitable for educational content. You can adjust the speed parameter to fit your script’s rhythm. These eight presets cover most narration needs without requiring custom voice design or cloning for your animation projects.

Are the voices gender-specific?

The eight voices are categorized as female (Aria, Nova, Luna, Sage) and male (Felix, Milo, Atlas, Orion). However, any voice can speak any of the 23 supported languages. You can adjust speed and text content to alter perception. There are no gender-specific constraints in the API, allowing flexible casting for your animated characters.

Can I adjust the voice tone?

You can adjust the speed parameter between 0.25 and 4.0 to change pacing. There are no emotion tags or SSML support for fine-tuning tone. The voices are based on licensed recordings from LibriTTS-R. You cannot upload custom pronunciation lexicons. To change tone, modify the input text or select a different voice preset from the eight available options.

What languages are best for these voices?

All eight voices can speak all 23 supported languages, including English, Spanish, Japanese, and German. If you do not set the language, the system guesses based on the script, defaulting to English for Latin text. For non-English Latin-script text, explicitly set the language parameter. Vietnamese, Indonesian, and Thai are not supported.

How do I test voices before buying?

Sign in with Google to receive $0.50 of free trial credit, valid for 7 days. This allows about 50,000 characters of generation. You can use the /v1/voices endpoint to check available voices and formats. Failed requests are free. Use this credit to test different voices and speeds before committing to a top-up via USDT.

Is there a demo for each voice?

The GET /v1/voices endpoint returns details on voices, languages, formats, and price. You can generate audio for each voice using the API to hear the quality. There is no built-in web demo player on the site. Use the free trial credit or a small top-up to generate samples for all eight voices directly.

Can I combine voices?

The API generates one audio file per request using a single voice preset. You cannot combine two voices in one request. To create a dialogue, generate each line separately with the desired voice and join the audio files. Split texts into parts of up to 4096 characters if needed, then merge the resulting files.

What is the difference between tts-1 and tts-1-hd?

Both models use the same engine and price. The only difference is the model identifier in the request. You can use either

Do voices sound different in different formats?

The audio content remains the same across mp3, wav, opus, flac, and pcm formats. The difference is in file compression and quality. Choose mp3 for smaller file sizes or wav for lossless quality. The voice characteristics do not change based on the response_format parameter. Select the format that best fits your animation pipeline.

Can I use voices in background music?

The API generates speech audio files. You can mix these files with your background music in post-production. There is no built-in music generation or mixing in the API. The generated audio carries an inaudible watermark marking it as synthetic. You are free to use the audio in your animations without additional licensing fees.

Female Voices Overview

Sound wave illustration for: Female Voices Overview

Vocaltoons provides three distinct female voices optimized for animation production, allowing creators to match character personalities precisely without licensing fees. Each voice offers a unique tonal quality suitable for different narrative roles within your storyboard.

Aria: Warm Narrator

Aria delivers a clear, warm tone ideal for narrator roles or friendly protagonist dialogue. Her vocal quality remains consistent across 23 languages, ensuring reliable character continuity.

  • Best for: Guiding narration, friendly hero dialogue
  • Tone: Warm, clear, approachable

Nova: Bright Upbeat

Nova offers a bright, energetic delivery perfect for upbeat scenes or younger characters. Her consistent pitch and pace help maintain high energy levels in promotional or lively segments.

  • Best for: Young characters, upbeat promos
  • Tone: Bright, energetic, lively

Luna: Soft Storytelling

Luna provides a soft, gentle voice suitable for storytelling or emotional scenes. Her calm delivery supports narrative depth without overpowering the visual elements of your animation.

  • Best for: Emotional scenes, bedtime stories
  • Tone: Soft, gentle, calming

Male Voices Overview

Sound wave illustration for: Male Voices Overview

The male voice lineup offers distinct tonal ranges for different narrative needs. Each preset is designed to deliver clear, consistent audio that integrates seamlessly into animation projects without requiring extensive post-processing.

These voices support all 23 available languages, allowing creators to maintain character consistency across diverse linguistic markets. The underlying engine ensures uniform quality regardless of the specific male preset selected.

Felix: Energetic Promo

Felix delivers high-energy, upbeat narration ideal for promotional content or dynamic scene transitions. His tone is bright and engaging, suitable for keeping viewer attention during fast-paced sequences.

Use Felix for voiceovers that require a sense of urgency or excitement, such as product highlights or action-oriented storytelling segments.

  • Best for: Promos, trailers, high-energy scenes
  • Tone: Bright, upbeat, engaging

Milo: Friendly Conversational

Milo provides a warm, approachable voice that mimics natural human conversation. This makes him ideal for dialogue-heavy scenes or characters that need to feel relatable and grounded.

His neutral yet friendly delivery works well for explanatory content or character interactions that require a sense of intimacy and trust.

  • Best for: Dialogue, explanations, relatable characters
  • Tone: Warm, approachable, natural

Atlas: Neutral News

Atlas offers a clear, authoritative, and neutral delivery perfect for news-style segments or factual narration. His voice is steady and professional, reducing listener fatigue during long exposés.

Use Atlas when clarity and objectivity are paramount, such as in documentary-style animations or informational sequences.

  • Best for: News, documentaries, factual narration
  • Tone: Clear, authoritative, steady

Orion: Deep Documentary

Orion provides a deep, resonant voice suited for serious, atmospheric, or historical narratives. His rich timbre adds gravity and depth to dramatic scenes or epic storytelling.

Ideal for voiceovers that require a sense of weight and presence, such as historical accounts or suspenseful reveals.

  • Best for: Dramas, history, suspense
  • Tone: Deep, resonant, atmospheric

Matching Voices to Character Types

Sound wave illustration for: Matching Voices to Character Types

Choosing the right voice is critical for animation consistency. Vocaltoons provides eight distinct presets, each designed for specific narrative roles. This guide maps these voices to character archetypes, ensuring your prototype sounds cohesive before final production begins.

By aligning voice traits with character personalities, you reduce iteration time during storyboarding. The following breakdown helps you select the optimal preset for protagonists, antagonists, and supporting roles.

Protagonist Voices

Protagonists require voices that convey relatability and emotional range. Aria offers a warm, clear narrator tone ideal for gentle leading roles. Luna’s soft, storytelling quality suits introspective heroes. Nova’s bright, upbeat delivery works for energetic, optimistic leads who drive the plot forward.

These voices maintain clarity over long dialogue sequences, ensuring the audience stays engaged with the main narrative arc without vocal fatigue.

  • Aria: Best for calm, wise leaders
  • Luna: Ideal for introspective or magical characters
  • Nova: Perfect for energetic, optimistic heroes

Antagonist Voices

Antagonists benefit from voices that command attention or evoke unease. Orion’s deep, documentary-style tone adds gravity and menace to villainous roles. Felix’s energetic, promo-like delivery can create a charismatic yet threatening villain, suitable for high-stakes conflicts.

Using a distinct vocal texture helps audiences instantly recognize the opposing force in your animation, enhancing dramatic tension without visual cues.

  • Orion: Best for serious, imposing villains
  • Felix: Ideal for charismatic or chaotic antagonists

Supporting Character Voices

Supporting characters need voices that complement the protagonist without overshadowing them. Sage’s calm, instructional tone works well for mentors or guides. Milo’s friendly, conversational voice suits relatable sidekicks. Atlas’s neutral, news-like delivery provides a stable background presence.

These presets ensure supporting roles remain distinct yet harmonious within the ensemble, maintaining audio balance during ensemble scenes.

  • Sage: Ideal for mentors or experts
  • Milo: Best for relatable sidekicks
  • Atlas: Suitable for neutral, background roles

Language Considerations

Sound wave illustration for: Language Considerations

Vocaltoons supports 23 languages, including Arabic, Chinese, French, German, Japanese, Spanish, and others. Any voice can speak any of these languages, ensuring consistent character tone across multilingual projects. The API detects language from the script if not specified, but Latin-script text defaults to English, so explicitly setting the language parameter is recommended for non-English content to avoid mispronunciation.

Supported languages are Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, and Turkish. Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, and Bengali are not supported. You can generate audio in any supported language using any of the eight voice presets, maintaining audio quality and consistency regardless of the language chosen.

English Voice Quality

All eight voices deliver natural, clear English speech optimized for animation dialogue. The system handles up to 4096 characters per request, making it suitable for long monologues or rapid-fire dialogue sequences common in cartoons.

For best results, ensure proper punctuation and spacing. The API counts spaces and punctuation as characters, so concise scripting helps manage your $0.01 per 1,000 characters budget efficiently.

  • Latin-script text defaults to English if language is unset
  • Explicitly set language for non-English Latin text
  • Supports full character set including punctuation and spaces

Non-English Voice Options

Each voice adapts to 23 languages, allowing you to maintain a consistent vocal style across different characters speaking different languages. For example, use 'Aria' for a narrator in both English and French scenes.

If the language field is omitted, the system guesses the language from the script. For non-English text in Latin script, always set the language parameter to ensure correct pronunciation and intonation.

  • 23 supported languages
  • Any voice can speak any supported language
  • Explicit language setting recommended for non-English Latin text

Accent Variations

Voices are based on licensed recordings from LibriTTS-R and LibriVox, providing natural, human-like speech patterns without artificial accent markers. There are no built-in accent variations; each voice has a consistent tone across all languages.

For example, 'Atlas' maintains a neutral, news-like delivery in English, Spanish, and German, while 'Orion' provides a deep, documentary-style tone in Japanese and Korean. This consistency helps maintain character identity across multilingual productions.

  • No artificial accent markers
  • Consistent tone across languages
  • Based on licensed LibriVox recordings

Technical Settings for Best Results

Sound wave illustration for: Technical Settings for Best Results

Optimizing technical settings ensures your animation audio matches the intended pacing and production quality. By adjusting speed, format, and model parameters, you can align the synthetic speech precisely with your storyboard timing and final render requirements.

These settings allow you to balance file size, compatibility, and voice fidelity without needing complex post-processing. Understanding how each parameter affects the output helps you maintain consistent character voices across episodes and scenes.

Speed Adjustment

You can set the speed between 0.25 and 4.0. This allows you to match the voice to the character’s energy level, from slow storytelling to fast-paced dialogue. Adjusting speed helps synchronize audio with animated lip movements without distorting the voice pitch unnaturally.

  • Range: 0.25 to 4.0
  • Preserves natural pitch
  • Syncs with animation timing

Format Selection

Choose from mp3, wav, opus, flac, or pcm. MP3 offers small file sizes for web delivery, while WAV or FLAC provides lossless quality for professional editing. PCM is useful for raw data integration. Selecting the right format balances storage needs with audio fidelity for your pipeline.

Failed requests do not consume credit, ensuring you only pay for usable audio files that meet your quality standards.

  • MP3 for web
  • WAV/FLAC for editing
  • PCM for raw data

Model Choice

Select either tts-1 or tts-1-hd. Both use the same engine and price, offering identical voice characteristics. The choice depends on your specific quality preference for background narration versus prominent dialogue. Both models support all 23 languages and 8 voice presets.

image]:

Cost Analysis by Voice

Sound wave illustration for: Cost Analysis by Voice

For animation studios, budget predictability matters. Vocaltoons charges $0.01 per 1,000 characters across all eight voices. Whether you use Aria for a warm narrator or Orion for a deep documentary, the rate is identical. This uniform pricing simplifies cost estimation for storyboards and prototypes.

Uniform Pricing

Each character’s dialogue incurs the same cost per character count. Spaces and punctuation count toward the total. Failed synthesis requests are free, ensuring you only pay for usable audio files delivered to your production pipeline.

  • $0.01 per 1,000 characters for all models
  • Failed requests are free
  • No monthly plans or minimum spend

Volume Discounts

Our pricing structure is flat, meaning there are no volume discounts. Every character, whether spoken by Felix or Sage, costs the same rate. This consistency helps startups forecast expenses for web series or educational content without complex tier calculations.

Trial Usage

New accounts signed in with Google receive $0.50 in trial credit. This covers approximately 50,000 characters, allowing you to test voice quality and integration before committing funds. Accounts created with email and password start without this trial credit.

  • $0.50 credit for Google sign-ups
  • Approximately 50,000 characters
  • Valid for 7 days

Integration with Animation Workflow

Sound wave illustration for: Integration with Animation Workflow

Vocaltoons integrates directly into your animation pipeline by converting script text into audio files via API, eliminating manual recording sessions. This allows storyboard artists to sync dialogue with visual drafts instantly, ensuring character timing aligns with visual cues before final rendering begins.

The workflow supports rapid iteration: update a line in your script, generate new audio, and replace the placeholder track. This loop continues until the emotional delivery matches the visual intent, preserving creative control without waiting for external voice actors.

Once approved, the high-fidelity audio files (MP3, WAV, or FLAC) are exported directly into your editing suite. The consistent voice quality across episodes ensures brand cohesion, while the API’s speed allows for last-minute script tweaks without production delays.

Storyboard to Audio

Generate placeholder dialogue early in pre-production to test pacing and visual timing. Use the speed parameter to adjust delivery rate, matching the character’s energy to the storyboard’s motion dynamics without re-recording.

Sync audio timestamps with keyframes. Since requests are processed sequentially, plan batch generation during off-peak hours to avoid queue delays, ensuring your storyboard review isn’t bottlenecked by API latency.

  • Use tts-1 for quick storyboard drafts; switch to tts-1-hd for final quality.
  • Set language explicitly to avoid English default for non-Latin scripts.
  • Export as WAV for lossless editing, or MP3 for faster previewing.

Iterative Editing

Refine dialogue delivery by adjusting the speed parameter between 0.25 and 4.0. This allows fine-tuning of emotional intensity without re-recording, enabling precise alignment with animated mouth movements and gestures.

Replace individual lines or entire scenes instantly. Since failed requests are free, you can experiment with different voices or speeds without penalty, ensuring the best fit for each character’s personality.

  • Test nova for upbeat scenes, sage for instructional segments.
  • Use response_format: pcm for raw audio processing.
  • Keep text under 4096 characters to ensure single-request generation.

Final Production

Export final audio files in your preferred format (MP3, WAV, OPUS, FLAC, or PCM). These files are ready for direct import into your animation software, maintaining consistency across all episodes or shorts.

Leverage the consistent voice profiles for all characters. Since each voice is based on licensed recordings, you avoid licensing disputes for background or recurring dialogue, streamlining your production budget and legal workflow.

  • Use tts-1-hd for final deliverables requiring high fidelity.
  • Store audio metadata for billing and audit trails.
  • No need for external voice actor contracts for consistent roles.
Integration with Animation Workflow

Common Mistakes to Avoid

Sound wave illustration for: Common Mistakes to Avoid

Animation production relies on consistent character voices, yet API misuse can break immersion. Avoid common pitfalls by aligning technical settings with creative needs. Proper configuration ensures your synthetic voices match the storyboard’s emotional tone without wasting credits on failed requests.

This section highlights frequent errors in tone selection, language handling, and request limits. Correcting these issues saves time during iterative editing and keeps your animation pipeline smooth.

Mismatched Tone

Using a voice that contradicts the character’s role disrupts audience engagement. For example, pairing Orion’s deep documentary tone with a bright, upbeat comedy scene creates cognitive dissonance. Ensure the voice preset aligns with the character’s personality archetype.

  • Match Aria or Luna for warm, narrative-driven protagonists.
  • Use Felix for energetic promo-style segments.
  • Avoid using Atlas for emotional, intimate scenes.

Ignoring Language

The API guesses language from Latin script, defaulting to English. If your character speaks Dutch or Swedish, omitting the language field may result in incorrect pronunciation. Always specify the ISO code for non-English languages to ensure accuracy.

  • Set language for any non-English dialogue.
  • Default assumption is English for Latin script.
  • Verify output for 23 supported languages.

Overloading Requests

Each request is limited to 4096 characters. Sending larger texts causes errors or truncation. Split long scripts into smaller chunks and join the audio files afterward. This prevents processing failures and maintains control over audio quality.

  • Limit input text to 4096 characters.
  • Split long scripts manually or via code.
  • Join audio files for final assembly.

Case Studies

Sound wave illustration for: Case Studies

Animation studios often face high costs for voice talent during the prototype phase. Vocaltoons provides a consistent, scalable alternative for generating dialogue across multiple characters without licensing fees.

By using our API, creators can iterate rapidly on script changes, ensuring character voices remain uniform throughout the production pipeline from storyboard to final render.

Short Film Example

A indie studio used the API to generate dialogue for a 5-minute animated short. They selected Aria for the narrator and Orion for the antagonist, maintaining a cohesive audio tone across 12 scenes.

The process took less than an hour, allowing the team to focus on visual animation rather than scheduling voice actors. Each character’s voice remained consistent even when scripts were tweaked late in production.

  • Generated 15,000 characters in under 10 minutes
  • Used Aria and Orion for distinct character voices
  • Saved ~$200 compared to hiring two voice actors

Web Series Example

A startup producing a weekly web series needed rapid turnaround for 10 episodes. They used Nova for the energetic protagonist and Milo for the supporting cast, enabling quick script revisions without re-recording.

The API’s speed adjustment allowed fine-tuning of pacing for comedic timing, ensuring each episode felt polished without the delay of traditional voice recording sessions.

  • Produced 10 episodes in 2 weeks
  • Used Nova and Milo for main characters
  • Adjusted speed to 1.2x for comedic timing

Educational Series Example

An educational platform used Sage for clear, instructional narration in a 20-part series. The calm, instructional tone of Sage ensured that complex information was delivered clearly to young audiences.

By using the API, the team could easily update content as curricula changed, maintaining voice consistency across all episodes without needing to re-record entire segments.

  • Used Sage for consistent instructional tone
  • Updated 20 episodes with minimal effort
  • Maintained clarity for young audiences

Voice Selection Checklist

Sound wave illustration for: Voice Selection Checklist

Selecting the right voice ensures your animated characters sound distinct and believable. Vocaltoons provides eight presets designed for specific narrative roles, from warm narrators to deep documentary styles. Use this checklist to align voice traits with your script's emotional tone and character arc.

Start by defining the personality of each role. Consider the target demographic and budget constraints to optimize your API usage. Proper selection reduces revision cycles and keeps production costs predictable.

Character Personality

Match voice timbre to character traits. Aria suits warm narrators, while Orion fits deep documentary roles. Felix works for energetic promos, and Milo offers a friendly conversational tone. Atlas provides a neutral news style, and Luna excels at soft storytelling. Nova adds bright energy for upbeat scenes.

  • Define core traits (e.g., calm, energetic).
  • Assign voice based on emotional range.
  • Test combinations for ensemble casts.

Target Audience

Consider language and accent needs. All eight voices support 23 languages, including English, Spanish, and Japanese. Ensure the voice matches the audience's linguistic expectations for clarity and engagement.

  • Identify primary language.
  • Verify voice support for that language.
  • Check clarity for target age group.

Budget Constraints

Pricing is uniform at $10 per 1 million characters. Faster speech reduces character count, lowering costs. Use the free trial credit of $0.50 to test voices before committing to large batches.

  • Calculate character count per episode.
  • Adjust speed to optimize length.
  • Monitor balance via dashboard.

Future Voice Updates

Sound wave illustration for: Future Voice Updates

Vocaltoons maintains a static set of eight licensed voices derived from LibriTTS-R and LibriVox recordings. We do not release new synthetic voices on a rolling schedule, nor do we offer custom voice cloning or design services. Your library remains consistent, ensuring character voices stay uniform across long-running animation series without unexpected changes or upgrades.

The voice pool is fixed at the eight presets described in our documentation. Any updates will relate to existing voice models rather than new additions. This stability supports reliable production pipelines for animation studios where voice consistency is critical for character identity.

New Voice Releases

We do not add new voices to the library. The eight available presets remain the complete set. This ensures your character voices stay consistent without requiring re-dubbing or asset updates when new models are introduced elsewhere in the industry.

Community Feedback

Feedback is collected via email support. While we do not guarantee that specific requests will result in new voices, input helps prioritize maintenance of the existing licensed recordings. There is no public roadmap or voting system for voice selection.

Licensing Changes

Our voices rely on LibriTTS-R (CC BY 4.0) and LibriVox public-domain recordings. If licensing terms for these source materials change, we will notify users via email. Such changes are unlikely but would affect attribution requirements rather than voice availability or quality.

Resources for Further Learning

Sound wave illustration for: Resources for Further Learning

The API documentation provides the definitive reference for all endpoints, request parameters, and response formats. It includes ready-to-use curl examples and explains the exact JSON structure required for the /v1/audio/speech endpoint, ensuring you can integrate Vocaltoons into your animation pipeline without guesswork.

Understanding the technical details helps you avoid common integration errors, such as incorrect authentication headers or unsupported response formats. The docs also clarify rate limits and character counting rules, which are critical for budgeting your production costs accurately.

API Documentation

Access the full technical reference covering all endpoints, including voice lists and model specifications. It details the JSON body structure for speech generation and explains how to handle errors or retries effectively.

The documentation also outlines the character counting logic, ensuring you know exactly how much credit each request consumes. This transparency helps you predict costs before sending your first audio file.

  • Complete endpoint reference
  • cURL integration examples
  • Error code definitions

Community Forums

While we do not offer a dedicated public forum, you can find community insights by searching for Vocaltoons integration tips in broader TTS communities. Many animation developers share their experiences with voice selection and speed adjustments in these open spaces.

Engaging with other developers can help you discover creative uses of the 8 preset voices, such as combining Felix for promos or Orion for documentaries, even if formal discussions happen elsewhere.

Support Channels

For direct assistance, support is available exclusively via email. This channel is ideal for specific questions about your account, billing, or API key issues. Since we do not offer phone support, email ensures a written record of your request.

Response times may vary during high load, but our team focuses on resolving technical and account-related issues efficiently. For general technical questions, the documentation remains the primary resource for immediate answers.

Listen to the 8 voices

Each sample below was generated by this API from the same sentence. Our studio transforms your startup's vision into captivating animated experiences that delight and engage your target audience.

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. LibriTTS-R paper (arXiv 2305.18802)
  4. FFmpeg documentation
  5. TRON developer hub: TRC-20
  6. W3C WAI: making audio and video accessible
  7. OpenAI API reference: create speech
  8. Node.js documentation
  9. IETF RFC 7845: Ogg Opus
  10. ISO 639 language codes
  11. Creative Commons Attribution 4.0
  12. openai-python on GitHub

More on this site