Text-to-Speech API for Language Learning Video Lessons

By SonicLingo editorial team · Updated 2026-11-19

This guide helps language learning creators select the right synthetic voices for video lessons using the SonicLingo API. We analyze eight distinct voice presets—from Sage’s calm instruction to Milo’s conversational tone—so you can match audio to your pedagogical goals. Each voice is based on licensed LibriTTS-R recordings, ensuring clear, consistent audio for thousands of practice tracks without hiring voice actors.

You will learn how to adjust speed (0.25–4.0) for different learner levels, choose between tts-1 and tts-1-hd for clarity, and maintain brand identity across 23 languages. This page provides specific recommendations for vocabulary, dialogue, and storytelling content, helping you build effective auditory learning materials with reliable, cost-effective API infrastructure.

Text-to-Speech API for Language Learning Video Lessons

Frequently asked questions

Which voice is best for beginners?

Aria is described as a warm, clear narrator, making it ideal for beginners. Sage offers a calm, instructional tone suitable for educational content. Both voices provide clear articulation for language learners. You can test these presets with the free trial credit to see which clarity level suits your specific lesson style best.

Can I use the same voice in all languages?

Yes, every voice can speak any of the 23 supported languages. You can use Aria for English lessons and the same voice for Spanish or Japanese tracks. If the language is not set, it guesses from the script, but you should explicitly set the ISO code for non-English Latin text to ensure accuracy.

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 name in your API request. You can choose either 'tts-1' or 'tts-1-hd' without any change in cost or generation speed. They produce identical audio quality.

Are the voices male or female?

The API offers four female voices: Aria, Nova, Luna, and Sage. There are also four male voices: Felix, Milo, Atlas, and Orion. You can select any of these eight presets based on your course’s tone. Each voice has a distinct character suitable for different educational contexts.

Can I change the voice mid-sentence?

No, each API request generates audio for a single voice preset. To change voices, you must send separate requests for each segment and join the audio files. The API does not support dynamic voice switching within a single audio stream or SSML tags for voice changes.

Do the voices sound robotic?

The voices are based on licensed recordings from LibriTTS-R and LibriVox, providing natural synthetic speech. Every file includes an inaudible watermark to mark it as synthetic. While not human, the tone is designed for clarity and consistency, avoiding the typical robotic sound of older TTS systems.

How do I pick a voice for my course?

Use the free trial credit to test different presets. Aria offers clear narration, while Milo provides a friendly, conversational tone. Atlas is neutral for news-style lessons, and Luna is soft for storytelling. Generate sample tracks to see which voice best matches your target audience’s learning preferences.

Is Sage good for news?

Sage is described as calm and instructional, making it suitable for educational content. For news, Atlas is the recommended voice as it is neutral and news-oriented. Sage works well for instructional segments, but Atlas provides a more standard news tone for current events or updates.

Can Luna do storytelling?

Yes, Luna is specifically described as soft and suited for storytelling. Her tone is ideal for narrative lessons or language stories. You can use her for longer texts to create an engaging atmosphere. Ensure the language parameter is set correctly to match the story’s language.

Does Milo sound friendly?

Yes, Milo is described as friendly and conversational. He is ideal for interactive lessons or direct address to students. His tone helps create a welcoming learning environment. You can use him for short phrases or longer instructional segments where a personal touch is beneficial.

Is Orion too deep for kids?

Orion is described as deep and suited for documentaries. While deep voices can be engaging, his tone might be less appealing to younger children compared to brighter voices like Nova or Aria. Test Orion with your specific audience to see if the depth suits your lesson style.

Can I use multiple voices in one video?

Yes, but you must generate each voice segment separately via API requests. Then, join the audio files using your video editing software. The API does not support multi-voice generation in a single request. This workflow allows you to create diverse characters for your language lessons.

How do I test voices before buying?

Sign up with Google to receive $0.50 of free trial credit, valid for 7 days. This allows about 50,000 characters of generation. Use this credit to test different voices and languages. Check your balance and usage on the /get-api-key/ page after signing in.

Are the voices licensed properly?

Yes, the voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain audiobook readings. This ensures proper licensing for your generated audio files. You can use these tracks in your video lessons without additional licensing concerns for the voice source.

Top Voices for Instructional Content

Sound wave illustration for: Top Voices for Instructional Content

SonicLingo’s API supports 23 languages, enabling consistent audio for global language learning creators. Selecting the right voice ensures instructional clarity without distracting the learner. Our presets are designed for specific pedagogical needs, from grammar explanations to conversational drills.

The following voices provide the structural foundation for educational video lessons. They offer distinct tonal qualities that align with different types of content, ensuring that synthetic speech remains engaging and professional across diverse curriculum materials.

Sage: Calm and Instructional Tone

Sage delivers a calm, instructional tone ideal for explaining grammar rules or procedural steps. Her steady delivery prevents cognitive overload during complex explanations, making her suitable for beginners who need clear, non-distracting audio cues.

Best for: Grammar rules, procedural instructions. Tone: Steady, clear, non-distracting. Languages: Supports all 23 available languages.

  • Ideal for step-by-step language drills
  • Reduces listener fatigue during long lessons
  • Consistent pacing for beginner comprehension

Atlas: Neutral News Style

Atlas provides a neutral, news-style voice that conveys information with authority and clarity. His tone is objective and professional, making it excellent for reading aloud textbooks, presenting facts, or delivering structured lesson summaries without emotional bias.

Best for: Fact-based lessons, textbook readings. Tone: Neutral, authoritative, clear. Languages: Supports all 23 available languages.

  • Excellent for formal vocabulary lists
  • Clear articulation for advanced learners
  • Neutral tone avoids cultural bias

Aria: Clear Narration Style

Aria offers a warm, clear narration style that brings stories and dialogues to life. Her voice is engaging yet professional, making it perfect for narrative-based lessons where context and emotion help reinforce memory retention.

Best for: Narrative lessons, story-based learning. Tone: Warm, clear, engaging. Languages: Supports all 23 available languages.

  • Enhances memory through narrative context
  • Warm tone keeps learners engaged
  • Clear enunciation aids pronunciation practice

Voices for Storytelling and Engagement

Sound wave illustration for: Voices for Storytelling and Engagement

Storytelling drives retention in language education. Luna and Nova provide distinct vocal styles for narrative-driven content, allowing creators to match audio tone precisely to lesson context.

These voices transform static text into engaging audio experiences, keeping learners attentive through varied pacing and vocal character without requiring human voice actors.

Luna: Soft Storytelling Voice

Luna delivers a soft, storytelling tone ideal for narrative vocabulary or cultural context lessons. Her gentle delivery supports immersive listening exercises where comprehension relies on emotional nuance.

Use Luna for long-form reading practice or historical anecdotes. Her consistent cadence helps learners parse complex sentence structures without auditory fatigue.

  • Ideal for narrative-driven lessons
  • Supports immersive listening exercises
  • Consistent, gentle cadence

Nova: Bright and Upbeat Tone

Nova offers a bright, upbeat tone perfect for high-energy drills or motivational content. Her energetic delivery maintains learner engagement during repetitive practice sessions.

Deploy Nova for quick vocabulary reviews or interactive quiz introductions. The upbeat quality signals active learning phases, distinguishing them from passive listening tasks.

  • High-energy drills
  • Motivational content
  • Interactive quiz introductions

Felix: Energetic Promo Style

Felix provides an energetic promo style suitable for lesson transitions or call-to-action segments. His dynamic voice captures attention between instructional blocks.

Use Felix for short announcements or segment introductions. His distinct tone helps learners identify structural shifts in the lesson flow.

  • Lesson transitions
  • Call-to-action segments
  • Distinct tonal shifts

Voices for Conversational Practice

Sound wave illustration for: Voices for Conversational Practice

Conversational practice requires natural pacing and clear articulation to help learners distinguish between words. Our API provides voices suitable for dialogue simulation, ensuring that practice tracks sound like real interactions rather than robotic readings.

These voices handle the nuance of everyday speech, making them ideal for building listening comprehension. You can generate thousands of unique dialogue pairs without hiring voice actors, maintaining consistency across your entire curriculum.

Milo: Friendly Conversational Tone

Milo offers a warm, approachable sound that mimics a friendly tutor or peer. His tone is relaxed yet articulate, perfect for casual dialogue exercises where learners need to feel comfortable practicing.

Use Milo for everyday conversation scenarios, such as ordering food or asking for directions. His natural inflection helps students recognize common phrasing and intonation patterns in real-world contexts.

  • Ideal for casual dialogue exercises
  • Natural inflection for everyday scenarios
  • Approachable tone for beginner learners

Orion: Deep Documentary Style

Orion provides a deeper, more authoritative voice suitable for structured conversational practice. His steady pacing helps learners focus on vocabulary and grammar without distraction from rapid speech.

This voice works well for instructional dialogues where clarity is paramount. It maintains a professional demeanor while remaining engaging enough for extended listening sessions.

  • Steady pacing for focus
  • Authoritative tone for structure
  • Professional demeanor for clarity

Matching Tone to Lesson Type

Align voice characteristics with your lesson objectives to enhance retention. Use Milo for high-frequency conversational phrases and Orion for more complex, structured dialogues.

Consistency in voice choice helps learners recognize different speaking styles within the same language. You can switch between voices to expose learners to diverse accents and tones.

  • Align voice with lesson objectives
  • Switch voices for diverse exposure
  • Maintain consistency within modules

Gender and Tone Diversity

Sound wave illustration for: Gender and Tone Diversity

SonicLingo provides eight distinct voice presets, evenly split between female and male options. This diversity allows language learning creators to maintain consistent brand identity while offering auditory variety. Each voice is engineered for specific instructional or narrative contexts, ensuring clarity and engagement across different lesson types.

4 Female Voice Options

Aria delivers a warm, clear narration style ideal for structured vocabulary lessons. Nova offers a bright, upbeat tone suitable for energetic drills. Luna provides a soft, storytelling voice for narrative exercises, while Sage ensures a calm, instructional tone for complex grammar explanations.

  • Aria: Warm, clear narrator
  • Nova: Bright, upbeat
  • Luna: Soft, storytelling
  • Sage: Calm, instructional

4 Male Voice Options

Felix brings an energetic, promo-style energy to quick reviews. Milo offers a friendly, conversational tone perfect for dialogue practice. Atlas provides a neutral, news-style delivery for objective content, and Orion delivers a deep, documentary style for immersive listening exercises.

  • Felix: Energetic, promo
  • Milo: Friendly, conversational
  • Atlas: Neutral, news
  • Orion: Deep, documentary

Balanced Representation

The API supports four female and four male voices, enabling balanced representation in audio materials. Creators can alternate genders to prevent listener fatigue or pair specific voices with character roles in stories. This diversity supports inclusive content without requiring additional voice actors.

Voice Consistency Across Languages

Sound wave illustration for: Voice Consistency Across Languages

SonicLingo’s API ensures that a voice remains consistent across all 23 supported languages. When you switch from English to French or Japanese, the timbre and pitch characteristics of Sage, Luna, or Atlas do not drift, providing a seamless listening experience for learners comparing pronunciation across languages.

This consistency is critical for educational brands. Students recognize the instructor’s voice regardless of the target language, reinforcing familiarity and reducing cognitive load during multi-lingual video lessons.

Same Voice, Different Languages

The TTS engine applies the same vocal synthesis model to every supported language. Whether generating text in Arabic, Korean, or Swedish, the voice retains its unique tonal fingerprint. This ensures that a learner hearing a lesson in German recognizes the exact same narrator who taught them Spanish earlier.

This uniformity eliminates the jarring effect of switching between different voice actors for different language modules. Your curriculum feels cohesive, as if taught by a single, multilingual instructor rather than a rotating cast of voices.

  • Supports 23 languages with consistent vocal traits
  • No drift in timbre or pitch across language switches
  • Ideal for comparative pronunciation exercises

Maintaining Brand Identity

Consistent voices build trust with your audience. When learners hear the same warm narration of Aria or the calm instruction of Sage across all your video lessons, they associate that specific tone with your brand’s educational standard.

Use one voice for core concepts and another for storytelling segments, but keep those choices consistent across all language versions of your course material. This auditory branding helps students navigate your content library intuitively.

  • Builds auditory recognition for your brand
  • Reinforces educational quality through consistent tone
  • Supports multi-course ecosystems seamlessly

Cross-Lingual Audio Projects

Educational platforms often produce parallel lessons in multiple languages. SonicLingo allows you to generate these in parallel while maintaining exact voice matching. This is essential for translation studies or comparative linguistics courses where voice consistency aids in isolating linguistic variables.

Use the API to generate side-by-side audio tracks in English and Spanish, ensuring the pacing and tone remain identical. This precision supports advanced language learning techniques like shadowing and contrastive analysis.

  • Generate parallel tracks with identical vocal traits
  • Supports comparative linguistics workflows
  • Ensures pacing matches across language versions

Technical Quality: HD vs Standard

Sound wave illustration for: Technical Quality: HD vs Standard

The API offers two model options for audio synthesis: tts-1 (Standard) and tts-1-hd (HD). Both utilize the same underlying engine and pricing, allowing creators to choose based on production needs without affecting cost efficiency. The selection is made via the model field in the JSON request body.

HD mode provides enhanced audio fidelity, which is particularly beneficial for language learners who need to distinguish subtle phonetic differences. Standard mode remains fully capable for general instruction, ensuring consistent quality across all lessons regardless of the selected model.

tts-1-hd Audio Clarity

The tts-1-hd model delivers higher fidelity audio compared to the standard tts-1 option. This increased clarity helps learners perceive nuanced pronunciation details in target languages, supporting better auditory comprehension and mimicry during practice sessions.

Both models support the same eight voice presets and 23 languages. The HD enhancement applies uniformly across all voices, ensuring that whether you use Sage for instructional content or Luna for storytelling, the audio remains crisp and professional.

  • Supports MP3, WAV, OPUS, FLAC, and PCM formats
  • Same character pricing as standard model
  • Ideal for phonetic distinction in language learning

Bandwidth Considerations

Higher fidelity audio files may require slightly more bandwidth, though modern networks handle this efficiently. For creators distributing content via video lessons or downloadable audio tracks, the difference in file size is minimal and rarely impacts user experience on standard connections.

When generating large batches of audio for course libraries, consider using compressed formats like MP3 or OPUS to optimize storage and delivery speeds without sacrificing the clarity benefits of the HD model.

  • MP3 and OPUS formats reduce file size
  • WAV/FLAC available for uncompressed needs
  • Negligible impact on standard video platforms

Learner Listening Experience

Clear audio reduces cognitive load for learners, allowing them to focus on language structure rather than straining to hear the speaker. This is critical for beginners who are still developing their auditory processing skills for new sounds and rhythms.

Consistent voice quality across lessons helps maintain a professional brand identity. Whether using the calm tone of Sage or the deep documentary style of Orion, the HD option ensures that every word is delivered with precision and clarity.

  • Reduces listener fatigue
  • Improves pronunciation mimicry
  • Supports diverse learning speeds

Speed Settings for Learner Levels

Sound wave illustration for: Speed Settings for Learner Levels

Adjusting playback speed allows educators to tailor audio difficulty to specific learner proficiency levels without altering the underlying voice character. This flexibility ensures that practice tracks remain accessible while maintaining natural intonation patterns essential for language acquisition.

The API supports precise speed control between 0.25x and 4.0x, enabling creators to generate differentiated content from a single script. This reduces production time while accommodating diverse classroom needs, from slow listening exercises for beginners to rapid-fire drills for advanced students.

Slower Speed for Beginners

Beginner learners benefit from reduced speeds (0.25–0.75x) which extend syllable duration, improving phoneme discrimination and comprehension. Slower audio gives students more processing time to identify word boundaries and stress patterns without distorting the natural rhythm of the speech.

This approach is particularly effective for vocabulary introduction tracks where clarity outweighs natural pacing. Educators can generate multiple speed variants of the same lesson to support progressive difficulty scaling within a single course module.

  • Enhances phoneme recognition for new learners
  • Reduces cognitive load during initial exposure
  • Supports differentiated instruction from one script

Faster Speed for Advanced Users

Advanced students often require 1.2x–2.0x playback to simulate real-world listening conditions and build processing speed. Faster tracks challenge learners to extract meaning from continuous speech, mirroring native speaker cadence and reducing reliance on explicit parsing.

Using accelerated audio helps bridge the gap between classroom learning and authentic media consumption. Creators can test listener comprehension thresholds by gradually increasing speed across lesson series.

  • Simulates native speaker cadence
  • Builds rapid processing skills
  • Bridges classroom and real-world audio

Adjustable Range 0.25–4.0

The API accepts speed values from 0.25 to 4.0, offering granular control for precise audio customization. Each request generates a unique audio file with the specified tempo, allowing for extensive A/B testing of learner engagement.

This wide range supports both extreme slowing for detailed pronunciation analysis and rapid playback for memory drills. Creators can dynamically adjust speeds per lesson type, ensuring optimal listening experiences across diverse proficiency levels.

  • Granular speed control per request
  • Supports 0.25x to 4.0x range
  • Enables precise A/B testing of audio
Speed Settings for Learner Levels

Voice Selection Based on Content Type

Sound wave illustration for: Voice Selection Based on Content Type

To generate effective language learning materials, matching voice characteristics to lesson objectives reduces cognitive load and improves retention. SonicLingo provides distinct presets optimized for specific pedagogical styles, allowing creators to maintain consistency across thousands of practice tracks without hiring multiple voice actors.

Vocabulary Lessons: Sage

Sage offers a calm, instructional tone ideal for presenting new terms and definitions. Its steady pacing helps learners focus on pronunciation without distraction. Use Sage for direct instruction modules where clarity is paramount.

  • Clear enunciation for accurate phonetic modeling
  • Consistent intonation reduces listener fatigue
  • Ideal for single-word or short-phrase drills

Dialogue Practice: Milo

Milo delivers a friendly, conversational tone that mimics natural speech patterns. This voice excels in role-play scenarios, making practice feel interactive rather than robotic. Use Milo to simulate real-world conversations between two speakers.

  • Natural rhythm supports turn-taking practice
  • Approachable tone encourages learner engagement
  • Suitable for A2–B1 level dialogue simulations

Reading Aloud: Luna

Luna provides a soft, storytelling voice perfect for extended texts or narratives. Its gentle delivery maintains listener interest during longer reading exercises. Use Luna for graded readers or cultural context lessons requiring sustained audio.

  • Smooth flow for paragraph-length content
  • Emotional nuance aids comprehension of tone
  • Best for C1+ advanced listening practice

Auditory Testing and Validation

Sound wave illustration for: Auditory Testing and Validation

Validating audio quality before scaling production ensures your language lessons resonate with students. SonicLingo’s API allows you to generate small batches of audio samples quickly, letting you verify pronunciation and tone without committing to large character counts.

You can test specific phonemes or difficult dialect nuances by generating short clips. This low-cost validation step helps maintain high standards for educational content before publishing full lesson tracks.

Sample Audio Generation

Use the API to generate short audio snippets for validation. Requests are processed sequentially, so under load a request may wait in a queue.

Send up to 4096 characters per request. For quick checks, generate 10–20 second clips to verify clarity and accent accuracy across your target languages.

  • Generate 5–10 sample clips per lesson variant
  • Test voice settings at 0.75x speed for beginners
  • Verify audio format compatibility with your LMS

Learner Feedback Loops

Share generated samples with a small group of target learners. Collect feedback on naturalness, clarity, and engagement levels before full deployment.

Use this data to refine voice selection or adjust speed settings for specific language pairs, ensuring the synthetic speech meets educational standards.

  • Gather feedback from 5–10 test learners
  • Rate clarity on a 1–5 scale
  • Identify preferred voices for specific age groups

A/B Testing Voices

Run parallel tests with different voices for the same text. Compare listener retention and comprehension rates to determine which voice works best for your audience.

This method helps you optimize engagement without hiring multiple voice actors, ensuring consistent audio quality across your entire course library.

  • Compare Sage vs. Luna for storytelling
  • Test Milo vs. Atlas for dialogue
  • Measure completion rates per voice variant

Licensing and Attribution

Sound wave illustration for: Licensing and Attribution

SonicLingo’s synthetic voices are derived from the LibriTTS-R dataset, which contains public-domain audiobook readings from LibriVox. This foundation ensures that the core vocal timbres used in our API are legally cleared for commercial use without complex royalty structures.

We provide clear attribution guidelines for creators who wish to acknowledge the source material, ensuring transparency in your educational content.

LibriTTS-R Source Material

The voice models are built upon the LibriTTS-R dataset, a high-quality collection of English audiobooks. Each preset voice is based on these licensed recordings, preserving the natural cadence and tone of the original narrators.

CC BY 4.0 Licensing

The underlying audio data from LibriVox is released under the Creative Commons Attribution 4.0 International license. This allows you to generate unlimited practice tracks for your language learning videos without worrying about per-track licensing fees or expiration dates.

  • Commercial use is permitted
  • Attribution is recommended
  • Modifications to the generated audio are allowed

Public Domain Origins

Many of the source readings come from public domain books, meaning the original text and performances are free from copyright restrictions. This reduces the risk of content ID claims on platforms like YouTube or Vimeo, keeping your lessons safe and accessible.

Combining Voices for Variety

Sound wave illustration for: Combining Voices for Variety

Creating varied audio tracks for language lessons often requires blending multiple voice profiles within a single lesson or across a course series. By alternating between male and female presets, you can distinguish between narrator segments and character dialogues, making instructional content more engaging without hiring additional voice actors.

Mixing Male and Female Voices

Use Sage or Atlas for clear instructional narration, then switch to Milo or Luna for conversational examples. This contrast helps learners distinguish between explanation and practice material. Since each voice has a distinct timbre, alternating them reduces listener fatigue during long vocabulary drills.

  • Use Sage for instructions, then Milo for dialogue examples.
  • Alternate between Atlas (news style) and Luna (storytelling) for reading practice.
  • Maintain consistent voice profiles across similar lesson types for brand recognition.

Segmental Voice Changes

Split longer texts into segments of up to 4,096 characters and assign different voices to each part. Combine the resulting audio files using a simple script. This approach allows you to create dynamic audio lessons where different characters or tones guide the learner through various exercises.

  • Split input text into 4,096-character chunks.
  • Assign different voices to each segment.
  • Combine audio files into a single track.

Dynamic Lesson Structures

Design lessons that adapt to learner level by adjusting voice speed and profile. Use slower speeds for beginners with clearer voices like Aria, and faster, energetic tones like Nova for advanced users. This flexibility ensures the audio matches the cognitive load of the content.

Monitoring Voice Performance

Sound wave illustration for: Monitoring Voice Performance

Track your API usage directly from the /get-api-key/ dashboard. Monitor character consumption against your balance to avoid unexpected pauses in lesson production.

Set up error monitoring to catch failed synthesis requests. Since failed requests are free, you can quickly identify and retry problematic inputs without cost penalties.

Usage Analytics Dashboard

Access real-time data on your API key usage. The dashboard displays current balance, characters consumed, and trial status, allowing you to plan large-scale lesson batches efficiently.

Review historical usage patterns to optimize your production workflow. Keep an eye on the 60 requests per minute limit to ensure smooth generation during peak times.

  • Real-time balance updates
  • Request limit tracking
  • Trial credit status

Character Count Tracking

Monitor exact character counts per request. Remember that spaces and punctuation count toward the 4,096 character limit per request.

Use the dashboard to estimate costs accurately. At $0.01 per 1,000 characters, precise tracking helps you budget for thousands of practice tracks.

  • Exact character counting
  • Cost estimation
  • Limit monitoring

Error Rate Monitoring

Track failed requests to identify common issues. Failed requests are refunded in full, so you can retry without financial loss.

Use error data to refine your text preprocessing. Ensure your input texts meet the 4,096 character limit and contain valid characters for optimal synthesis.

  • Failed request tracking
  • Refund verification
  • Input validation tips

Listen to the 8 voices

Each sample below was generated by this API from the same sentence. This is critical for beginners who are still developing their auditory processing skills for new sounds and rhythms.

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. MDN: Web Speech API
  3. LibriVox public-domain audiobooks
  4. FFmpeg documentation
  5. IETF RFC 7845: Ogg Opus
  6. LibriTTS-R corpus (OpenSLR 141)
  7. openai-python on GitHub
  8. Python documentation
  9. openai-node on GitHub
  10. Node.js documentation
  11. TRON developer hub: TRC-20
  12. curl documentation

More on this site