Transparent Pricing for Text-to-Speech API Usage

By EchoWave editorial team · Updated 2026-11-19

This page details the transparent, character-based pricing for the EchoWave TTS API, designed specifically for indie developers and streamers evaluating synthetic voice solutions. We break down the standard rate of $10 per million characters, explaining how spaces and punctuation contribute to usage costs without hidden fees or monthly subscriptions.

You will find clear information on USDT top-ups via TRON, free trial eligibility, and technical limits affecting your budget. Whether you are generating 1-hour stream audio or short sound effects, this guide helps you calculate exact costs and understand the billing mechanics, rate limits, and voice options available for your project.

Transparent Pricing for Text-to-Speech API Usage

Frequently asked questions

How much does it cost to generate 1 minute of audio?

Cost depends on text length, not duration. The rate is $0.01 per 1,000 characters. Since generation is roughly real-time (25 characters per second), one minute of speech contains approximately 1,500 characters. Therefore, generating one minute of audio requires approximately 1,500 characters (at 25 characters per second). This results in a cost of $0.015 per minute. Spaces and punctuation count toward the character total.

Does the price change based on the voice selected?

No, the price is consistent across all voices. Both the 'tts-1' and 'tts-1-hd' models use the same engine and pricing structure. You can choose from any of the eight presets—such as Aria, Nova, or Felix—without any variation in cost per character.

Are there any fees for using different audio formats?

No additional fees apply for format selection. You can choose from mp3, wav, opus, flac, or pcm. The cost remains $0.01 per 1,000 characters regardless of the output format. Each format serves different bandwidth and quality needs without affecting your account balance.

How long does the free trial last?

The free trial lasts for 7 days. It is available when you sign in using Google. You receive $0.50 of credit, allowing approximately 50,000 characters of generation. Accounts created with email and password do not receive this trial credit automatically.

Can I top up with Bitcoin or Ethereum?

No, only USDT on the TRON network (TRC-20) is accepted. We do not support Bitcoin, Ethereum, credit cards, or PayPal. Payments are processed directly via blockchain transfer. Ensure you send the correct token on the correct network to receive automatic credit.

What is the minimum amount to top up?

The minimum top-up amount is $5. There is no maximum limit, and no monthly plans or subscriptions are required. Once the transfer is confirmed on the TRON network, your balance is credited automatically within minutes, ready for immediate use.

Do I lose my balance if I don't use it?

No, your balance does not expire. Funds added via USDT top-up remain in your account indefinitely until used. There are no monthly fees or inactivity charges. You can top up whenever needed, and the credit will be available for future API requests.

Are failed requests charged?

No, failed requests are free. If synthesis fails for any reason, the charge is returned in full to your account balance. You only pay for successfully generated audio files. This ensures you are not penalized for temporary network or processing issues.

Can I use the API for commercial projects?

Yes, the API is suitable for commercial use. Voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain readings. Generated audio files carry an inaudible watermark identifying them as synthetic speech.

What happens if I exceed the rate limit?

You are limited to 60 requests per minute per key. Before the first top-up, the limit is 10 requests per minute. If you exceed the limit, requests may queue or return an error. Splitting long texts into smaller parts helps manage throughput effectively.

Is there a limit to the length of text I can send?

Each request supports up to 4,096 characters. For longer texts, you must split the input into parts and join the resulting audio files. This approach allows you to generate extended content while staying within the single-request character limit constraints of the API.

Can I change my API key if it gets compromised?

Yes, you can issue a new key at any time. When you generate a new key, the old one stops working immediately. This ensures that if your key is exposed, you can revoke access instantly without waiting for a manual review or support ticket.

Do you support SSML for better control?

No, SSML is not supported. The API accepts plain text input only. You cannot use SSML tags for pronunciation or prosody control. If you need specific pacing, you can adjust the speed parameter between 0.25 and 4.0 to modify the output rate.

Is there a discount for high-volume users?

No, there are no volume discounts or tiered pricing. The rate remains $0.01 per 1,000 characters regardless of usage volume. There are no monthly plans or minimum spend requirements. You pay only for the characters you generate, with no additional fees.

How do I know which voice to choose?

Use the GET /v1/voices endpoint to list all available voices and their characteristics. You have eight presets: Aria, Nova, Luna, Sage, Felix, Milo, Atlas, and Orion. Each voice has a distinct tone, from warm narration to deep documentary styles, suitable for different content types.

Can I use the audio files offline?

Yes, generated audio files are returned in the response and can be stored locally. The API does not store the audio after delivery. You can download and use the files offline for streams, videos, or applications without any ongoing connectivity requirements.

Do you offer a refund if I am unsatisfied?

Failed requests are credited back automatically. If you are unhappy with the output quality, you can stop using the service. There are no subscriptions to cancel, so you simply stop making requests.

Is my data secure when I send text to the API?

Input text is used only to produce audio and is not used for training. Generated audio is returned in the response and not stored on our servers. Request metadata like time and status is kept for billing purposes only, ensuring privacy for your content.

Can I integrate this API with my WordPress site?

No, we do not offer an official WordPress plugin. However, you can integrate the API using the OpenAI SDK or plain HTTP requests. Since the endpoint shape matches the OpenAI speech endpoint, you can adapt existing integrations by changing the base URL to our API.

Is there a mobile app for managing my account?

No, there is no official mobile app. You manage your account, view balance, and generate keys via the website at /get-api-key/. You can access the documentation and API endpoints from any device with a web browser and internet connection.

Cost calculator

Estimated cost: $2.50

Spaces and punctuation count as characters. $10 per 1 million characters. Failed requests are free.

Standard Rate Structure

Sound wave illustration for: Standard Rate Structure

Our pricing model is straightforward: you pay $10 per 1 million characters processed. This flat rate applies regardless of the voice selected or the audio format chosen, ensuring predictable costs for your projects.

$10 per 1 million characters

The base cost is $10 for every million characters processed, equating to $0.01 per 1,000 characters. This rate remains consistent whether you use the standard "tts-1" or the high-definition "tts-1-hd" model, as both engines share the same pricing structure.

Spaces and punctuation count as characters

Character counting includes all spaces, punctuation marks, and special characters within your input text. For example, a 4,096-character limit per request includes the trailing spaces and punctuation, so plan your text splitting accordingly to avoid truncation.

No minimum spend or monthly plans

There are no monthly subscriptions or required minimum spending thresholds. You only pay for what you generate, making it ideal for indie developers with variable usage patterns. Unused credits from top-ups remain in your account indefinitely.

Free Trial Details

Sound wave illustration for: Free Trial Details

New users signing in via Google receive an immediate $0.50 credit, providing approximately 50,000 characters of free trial usage. This allows indie developers and streamers to test voice quality and API integration without any upfront payment.

The trial credit remains active for 7 days from account creation. If the balance is not exhausted within this window, the remaining amount expires and does not carry over.

$0.50 credit for new Google sign-ins

Upon authenticating with Google, your account is instantly credited with $0.50. This equates to roughly 50,000 characters, sufficient for generating short dialogues or testing voice presets across different characters.

This credit is applied automatically and appears in your dashboard balance. You can use it to generate audio files just like a paid account, allowing you to verify the API endpoint behavior before funding your account.

  • Instant credit upon Google sign-in
  • Approximately 50,000 characters available
  • No credit card required to start

Trial valid for 7 days

The trial period lasts exactly 7 days from the moment your account is created. Any unused credit expires automatically after this period and does not roll over to future billing cycles.

This short window is designed to let you evaluate latency, voice clarity, and integration ease quickly. If the service meets your needs for VTuber streams or game audio, you can proceed to top up.

Email/password accounts start with zero credit

Accounts created using a standard email and password registration do not receive the $0.50 trial credit. You must perform a manual top-up using USDT to begin generating audio.

This distinction ensures that trial users are identified via the Google sign-in flow. Email-based accounts are ready for immediate commercial use once funded, with no trial expiration concerns.

Payment Methods

Sound wave illustration for: Payment Methods

We accept USDT on the TRON network (TRC-20) for all top-ups. This streamlined payment method eliminates traditional banking delays and credit card processing fees, ensuring your account is funded quickly and efficiently. There are no invoices, VAT charges, or subscription management complexities to worry about.

USDT on TRON network (TRC-20)

Payments are processed exclusively via the TRON blockchain using the TRC-20 standard. This ensures low transaction fees and fast confirmation times compared to traditional fiat methods. We do not accept credit cards, PayPal, or other external payment processors.

Minimum top-up of $5

The minimum top-up amount is $5. This low threshold makes it accessible for indie developers and streamers to start using the API without a significant upfront commitment. You can add small amounts as needed to match your project's cash flow.

Automatic credit within minutes

Once your USDT transfer is confirmed on the network, your balance is updated automatically within minutes. There is no manual review or waiting period for support staff. Your available characters are immediately usable for API requests.

Cost Examples for Streamers

Sound wave illustration for: Cost Examples for Streamers

Understanding audio generation costs helps indie developers and streamers budget their synthetic voice usage effectively. With a flat rate of $0.01 per 1,000 characters, expenses scale directly with content length, making it easy to predict monthly API spend without tiered complexities.

Cost for a 1-hour stream at normal speed

Generating one hour of audio at normal speed (roughly real-time, ~25 characters/second) requires approximately 90,000 characters. At the standard rate, this costs $4.50. This predictable cost structure allows streamers to allocate budget for continuous background narration or character dialogue without hidden fees.

Cost for short sound effects and alerts

Short phrases or alert sounds typically contain fewer characters, taking a few seconds to generate. Even if a request fails, the charge is returned in full. For frequent, short interactions, the low character count keeps individual costs negligible, allowing for high-frequency usage within the 60 requests/minute limit.

Cost for long-form video voiceovers

For long-form content, split texts into parts of up to 4096 characters to stay within request limits. Joining the resulting audio files is straightforward. Since longer texts take time proportional to their length, budgeting remains simple: multiply total characters by $0.00001. No volume discounts or per-voice extra costs apply.

Billing and Refunds

Sound wave illustration for: Billing and Refunds

Understanding the financial mechanics of our TTS API is straightforward. We ensure that you only pay for successful synthesis, while our billing system tracks usage for accurate invoicing.

Failed requests are free

If a synthesis fails due to server issues or invalid parameters, no charge is applied. You are billed only when the audio file is successfully generated and returned.

Charge returned in full on error

In the rare event that a charge is processed for a failed request, the amount is returned to your account balance in full. This ensures your credit remains intact regardless of transient technical errors.

Balance does not expire

Your account balance is valid indefinitely. There is no expiration date for your credits, allowing you to top up when convenient and use the funds at your own pace for future streaming or development needs.

No Hidden Fees

Sound wave illustration for: No Hidden Fees

Our pricing model is designed for transparency, ensuring you pay exactly for what you consume without unexpected deductions. The flat rate of $0.01 per 1,000 characters applies uniformly, regardless of the voice or format selected. There are no hidden surcharges for using specific audio outputs like WAV or OPUS, nor extra fees for high-speed generation.

We avoid traditional payment overheads entirely. Since we accept USDT (TRC-20) exclusively, there are no credit card processing fees or merchant service charges passed on to you. Additionally, we do not charge VAT or invoicing fees, keeping your costs predictable and simple for indie developers and streamers budgeting their production expenses.

No VAT or invoicing fees

We operate without the administrative burden of VAT handling or complex invoicing systems. Your balance reflects the true cost of your API usage without regional tax additions or document fees.

This simplicity is particularly beneficial for international developers who avoid cross-border tax complexities. You can start generating audio immediately without waiting for invoice approvals or dealing with currency conversion fees from payment processors.

  • No VAT added to your balance
  • No invoicing processing fees
  • Simple, flat-rate billing

No credit card processing charges

By using USDT on the TRON network, we avoid credit card processing fees. This means more of your top-up value goes directly into your API balance. The minimum top-up of $5 ensures efficiency without sacrificing liquidity for smaller projects.

Unlike platforms that inflate prices to cover merchant fees, our rates remain consistent. You pay the base character rate without hidden markups designed to cover payment gateway costs.

No per-voice or per-format extra costs

All 8 voice presets (Aria, Nova, Luna, Sage, Felix, Milo, Atlas, Orion) cost the same per character. Whether you generate audio in MP3, WAV, OPUS, FLAC, or PCM, there are no format-specific premiums. The price is strictly tied to the input text length.

Speed adjustments (0.25x to 4.0x) also do not affect the base character cost. You only pay for the text processed, making it easy to calculate costs for long-form voiceovers or short stream alerts without complex pricing tiers.

Volume Discounts

Sound wave illustration for: Volume Discounts

EchoWave applies a single flat rate of $10 per 1 million characters regardless of how much you generate. Whether you process 1,000 or 1,000,000 characters, the cost per character remains exactly $0.01 per 1,000 characters. There are no volume-based reductions or bulk purchase incentives.

Flat rate regardless of volume

Your pricing does not change based on usage quantity. Every character, including spaces and punctuation, is billed at the same standard rate. This simplicity means you can scale your TTS usage without recalculating costs or negotiating tiered rates as your project grows.

No tiered pricing structures

We do not use complex pricing tiers that complicate budgeting. You do not pay less per character for higher volumes, nor do you pay more for exceeding a threshold. The structure is linear and predictable, making it easy to estimate costs for long-form content or high-frequency streaming alerts.

Consistent cost for indie developers

Indie developers benefit from this uniform pricing model. You can predict expenses for a single video voiceover or a full stream’s worth of character dialogue without worrying about hidden volume discounts or penalties. This consistency supports tight budgets common in indie game development and streaming setups.

Technical Limits Affecting Cost

Sound wave illustration for: Technical Limits Affecting Cost

Understanding technical constraints is essential for accurate cost estimation and reliable integration. The API enforces specific limits on input length, request volume, and processing speed that directly influence how you structure your audio generation workflows.

4096 character limit per request

Each API call accepts a maximum of 4096 characters, including spaces and punctuation. If your script or narration exceeds this threshold, you must split the text into chunks and concatenate the resulting audio files. This ensures consistent processing but requires client-side logic for long-form content.

  • Maximum input: 4096 characters per request
  • Spaces and punctuation count toward the limit
  • Multiple requests needed for longer texts

Splitting long texts for audio generation

For texts exceeding 4096 characters, split the input into manageable segments and join the returned audio files. Since requests are processed sequentially, long texts may experience queuing delays. Typical generation speed is roughly real-time, averaging 25 characters per second, meaning longer splits take proportionally longer to synthesize.

  • Split text into ≤4096 character chunks
  • Concatenate returned audio files manually
  • Processing is sequential; queueing may occur

Rate limits: 60 requests per minute

Each API key is limited to 60 requests per minute, or 10 per minute for new accounts before the first top-up. Exceeding this limit may result in temporary throttling. For streaming or high-frequency alert systems, implement exponential backoff or batch processing to stay within these bounds without incurring unnecessary wait times.

  • Limit: 60 requests per minute per key
  • New accounts: 10 requests per minute initially
  • Implement backoff strategies for high volume

Account & Security

Sound wave illustration for: Account & Security

Access to the EchoWave TTS API is secured through a single active API key per account, ensuring clear attribution and simplified billing tracking. This design minimizes the attack surface while keeping integration straightforward for indie developers and streamers who need reliable, auditable access to synthetic voice generation.

One active key per account

Each account maintains exactly one valid API key at any given time. This key is used in the Authorization: Bearer header for all requests. If you distribute the key across multiple services, any usage counts against your single account balance and rate limits.

  • Single key per account simplifies tracking
  • Used for all API requests
  • No concurrent key versions

Issue new key to invalidate old one

You can generate a new key at any time via the dashboard. The previous key stops working immediately upon issuance, allowing you to revoke access instantly if a key is compromised or no longer needed. This ensures rapid security response without waiting for expiration.

  • Instant invalidation of old key
  • Generate new key anytime
  • Immediate security revocation

Google or email/password sign-up

Accounts are created either via Google sign-in or standard email/password registration. Google sign-ups receive a $0.50 free trial credit valid for 7 days, while email/password accounts start with zero credit. Both methods provide equal access to the full API capabilities and billing features.

  • Google sign-in with trial credit
  • Email/password option available
  • Equal feature access

Supported Languages & Voices

Sound wave illustration for: Supported Languages & Voices

EchoWave supports 23 languages, including English, Spanish, Japanese, and Arabic, ensuring broad coverage for indie developers and streamers. Latin-script text defaults to English, but the language header allows precise control for other supported tongues.

The catalog features 8 distinct voice presets: Aria, Nova, Luna, and Sage (female); Felix, Milo, Atlas, and Orion (male). Each voice is trained on licensed recordings from LibriTTS-R and LibriVox, offering clear, distinct tones suitable for VTuber avatars or game narration.

23 languages supported

We support 23 languages: Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, and Turkish. If you send non-English Latin text without specifying a language, we default to English.

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

8 voice presets available

Our eight voices cover a wide range of styles. Female options include Aria (warm narrator), Nova (bright/upbeat), Luna (soft storytelling), and Sage (calm/instructional). Male options feature Felix (energetic promo), Milo (friendly/conversational), Atlas (neutral/news), and Orion (deep/documentary).

  • Aria, Nova, Luna, Sage (Female)
  • Felix, Milo, Atlas, Orion (Male)

Any voice can speak any supported language

Unlike some competitors, our voices are not locked to specific languages. Any of the 8 presets can speak any of the 23 supported languages. This flexibility allows you to maintain a consistent character voice across different linguistic markets without switching models.

Audio Formats & Quality

Sound wave illustration for: Audio Formats & Quality

The API supports five audio formats to suit different streaming and development needs. You can choose between MP3, WAV, OPUS, FLAC, or PCM. Each format is generated on demand with no extra fees for format selection. This flexibility allows you to optimize file sizes for streaming or preserve lossless quality for post-production.

Speed control ranges from 0.25x to 4.0x, enabling precise pacing adjustments for your content. Whether you need a slow, narrative tone or a rapid-fire delivery for alerts, the engine adapts instantly. The default speed is 1.0x, which typically generates audio at roughly real-time speed.

MP3, WAV, OPUS, FLAC, PCM formats

Select your preferred output format directly in the request. MP3 offers high compression. WAV and FLAC provide lossless quality for professional editing. PCM is available for raw data processing. OPUS is ideal for efficient bandwidth usage. There are no per-format surcharges.

  • MP3: Compressed, suitable for streaming
  • WAV: Lossless, standard for editing
  • OPUS: Efficient bandwidth usage
  • FLAC: Lossless, archival quality
  • PCM: Raw, uncompressed data

Speed control from 0.25x to 4.0x

Adjust playback speed to fit your specific use case. Lower speeds enhance clarity for instructional content. Higher speeds work well for short alerts or rapid updates. The engine maintains voice consistency across all speed settings. This feature is included in the standard character count.

  • Range: 0.25x to 4.0x
  • Maintains voice consistency
  • No additional cost for speed changes
  • Optimize for streaming or editing

Inaudible watermark on all files

Every generated audio file contains an inaudible watermark. This marks the file as synthetic speech. The watermark is embedded automatically. It does not affect audio quality. This feature helps identify generated content without requiring visual metadata.

  • Inaudible synthetic marker
  • Embedded automatically
  • No quality impact
  • Identifies AI-generated audio

Data Privacy & Usage

Sound wave illustration for: Data Privacy & Usage

Your input text is used exclusively for the immediate synthesis of audio and is not retained for model training. This ensures that your unique scripts, character dialogues, or stream content remain private and are not repurposed to improve the underlying TTS engine.

Generated audio files are returned in the response and are not stored on our servers after delivery. Once you receive the audio, our infrastructure does not keep a copy, reducing the risk of data leakage or unauthorized access to your final content.

Input text not used for training

We process your text solely to generate the requested audio file. Unlike many consumer-grade TTS services, we do not harvest your input data to train or refine our voice models. Your scripts remain yours.

This approach is critical for indie developers and streamers who generate proprietary content. Whether you are creating VTuber dialogue or narrating a long-form video, your text input is transient and not added to our training dataset.

  • Text is used only for immediate synthesis
  • No data mining for model training
  • Proprietary scripts remain private

Generated audio not stored

Once the audio file is transmitted to your client, it is not permanently stored on our servers. We do not maintain a library of your generated sounds, ensuring that your unique voice outputs are not accessible to us or third parties after delivery.

This ephemeral storage model aligns with the needs of streamers who generate dynamic alerts or sound effects on the fly. You retain full control over the generated files without worrying about them lingering in our database.

  • Audio is returned in the response
  • No permanent storage of generated files
  • Immediate deletion after delivery

Metadata kept for billing only

We retain minimal metadata including the timestamp, character count, and request status. This information is strictly used for accurate billing, rate limit tracking, and troubleshooting service issues.

This data does not include the content of your text or the resulting audio. It is a lightweight record necessary to ensure you are charged correctly for the characters processed and to monitor API usage limits.

  • Timestamps and character counts
  • Request status for billing
  • No content metadata retained

Support & Availability

Sound wave illustration for: Support & Availability

We offer straightforward support via email. Our team reviews inquiries and provides technical assistance without the complexity of tiered support levels. This approach keeps our overhead low, which helps maintain our transparent, per-character pricing model for indie developers and streamers.

Email support only

All support requests are handled through our dedicated email channel. We prioritize clear, direct answers to your technical questions regarding integration, voice selection, and billing. This ensures that every inquiry receives full attention from our engineering team without being routed through a general support queue.

  • Direct access to technical support via email
  • No automated ticketing systems to navigate
  • Clear, written records of all support interactions

No live chat or phone number

We do not offer live chat widgets or a phone support line. This decision allows us to focus resources on API stability and cost-efficiency rather than maintaining large support teams. For indie developers and streamers, email provides a reliable, asynchronous way to resolve issues without waiting on hold or dealing with chat bots.

No SLA or uptime guarantee

Our service operates without a formal Service Level Agreement (SLA) or guaranteed uptime percentage. While we strive for consistent performance, we do not offer financial credits for downtime. This model suits developers who prioritize low cost and flexibility over enterprise-grade uptime guarantees, allowing us to keep prices competitive at $0.01 per 1,000 characters.

Support & Availability

SDK & Integration

Sound wave illustration for: SDK & Integration

Our API is designed to integrate seamlessly with existing developer workflows. By simply adjusting the base URL, you can use the official OpenAI SDKs to interact with our text-to-speech endpoint without learning a new client library.

OpenAI SDK compatible

The request structure mirrors the standard OpenAI speech endpoint, allowing you to send JSON payloads with identical field names like model, input, and voice. This compatibility means you can swap the base URL and continue using your preferred language SDKs for Python, JavaScript, or Go without rewriting your integration logic.

  • Supports standard JSON request bodies
  • Compatible with official OpenAI SDKs
  • Works with Python, JS, Go, and other languages

Change base_url to use our API

To start using our service, update your SDK configuration to point to https://api.ktvsuper.info/v1. All other parameters, including authentication headers and request bodies, remain the same as you would use for the OpenAI service. This allows for a quick drop-in replacement in your application.

No official custom SDK provided

We do not offer a proprietary SDK or mobile-specific library. We rely on the robust, widely-used OpenAI SDKs, which are actively maintained and supported by large communities. This ensures stability and reduces the need for us to maintain separate codebases for different platforms.

Common Misconceptions

Sound wave illustration for: Common Misconceptions

EchoWave is an independent text-to-speech API provider, not a subsidiary or partner of OpenAI. While our endpoint structure mirrors OpenAI’s speech endpoint for easy SDK migration, we operate separately with our own infrastructure, pricing, and voice engine.

Not affiliated with OpenAI

We use a compatible request shape, but our models and voices are distinct. You are not using OpenAI’s TTS service; you are using EchoWave’s engine, which processes requests independently.

  • Independent infrastructure
  • Separate billing and keys
  • Distinct voice engine

No voice cloning available

Custom voice cloning is not offered. You cannot upload a sample to create a new voice. All 8 presets (Aria, Nova, Luna, Sage, Felix, Milo, Atlas, Orion) are based on licensed recordings from LibriTTS-R and LibriVox.

  • No custom voice uploads
  • 8 fixed presets only
  • Licensed source material

No SSML or emotion tags

The API does not support SSML, phoneme dictionaries, or emotion tags. Control is limited to speed (0.25–4.0x) and voice selection. Complex prosody must be handled by adjusting text input or using external tools.

  • No SSML support
  • No emotion tags
  • Speed control only

Getting Started

Sound wave illustration for: Getting Started

Begin by visiting the /get-api-key/ page on ktvsuper.info and signing up via Google or email. This creates your independent account and generates a unique Bearer token for authenticating your requests.

Sign up to get your API key

Use the provided curl example to verify your credentials immediately. Send a POST request to the speech endpoint with a short test string to confirm connectivity and basic synthesis capabilities.

  • Sign up via Google or email/password
  • Receive a unique Bearer token
  • Authenticate with Authorization header

Use the provided curl example

Replace the base URL in your OpenAI SDK or HTTP client to point to https://api.ktvsuper.info/v1. This allows you to use familiar SDK methods while accessing our TTS engine.

Check balance and usage dashboard

Monitor your remaining characters and billing status directly on the account page. Track your usage history and ensure your USDT top-ups are credited correctly within minutes.

Troubleshooting Payments

Sound wave illustration for: Troubleshooting Payments

When topping up your EchoWave account, ensure you send USDT specifically on the TRON (TRC-20) network. Transfers using other networks or tokens may not be detected automatically, delaying your credit. The system processes incoming transactions automatically, typically crediting your balance within minutes of network confirmation.

USDT transfer confirmation time

Confirmation speed depends on the TRON network latency. Once the transaction is confirmed on-chain, your balance updates automatically without manual intervention.

  • Network-dependent latency
  • Automatic credit update
  • TRC-20 only

Minimum $5 top-up requirement

The minimum top-up amount is $5. This ensures efficient processing of the blockchain transaction relative to the value added. Smaller amounts may be economically inefficient for the automatic crediting system.

Balance does not expire

Your credit balance remains active indefinitely. There is no expiration date for your funds, allowing you to top up when convenient and use credits at your own pace.

Voice Selection Guide

Sound wave illustration for: Voice Selection Guide

Choosing the right voice enhances your stream’s immersion. Our eight presets cover distinct tonal ranges, from warm narration to energetic promos.

Each voice is designed for specific content types. Selecting the correct preset ensures your synthetic audio matches the mood of your VTuber character or stream segment without requiring additional editing.

Female voices: Aria, Nova, Luna, Sage

Aria offers a warm, clear narrator tone suitable for storytelling. Nova provides a bright, upbeat energy ideal for lively streams. Luna delivers a soft, storytelling voice for calming segments. Sage is calm and instructional, perfect for tutorials or guides.

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

Male voices: Felix, Milo, Atlas, Orion

Felix is energetic and promo-focused, great for announcements. Milo sounds friendly and conversational, engaging viewers naturally. Atlas provides a neutral, news-style delivery for updates. Orion offers a deep, documentary tone for serious content.

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

Choosing voice for content type

Match the voice to your stream’s genre. Use Felix for high-energy moments and Milo for chat interactions. Select Luna or Aria for narrative-driven VTuber roles. Choose Atlas or Orion for structured, informational segments.

image":

Language Support Details

Sound wave illustration for: Language Support Details

The API automatically detects the language from the text script, defaulting to English for any Latin-script input. This simplifies integration for English-first projects, but requires explicit configuration for other languages to ensure accurate synthesis.

For non-English content, you must set the language parameter to the ISO code. Without this, the system assumes English, potentially mispronouncing words in other Latin-based languages like Spanish or French.

Latin script defaults to English

If the language field is omitted, the engine assumes English for all Latin characters. This works perfectly for English text but may cause awkward pronunciation for other languages using the same script.

Set language header for non-English

Specify the language parameter (e.g., fr, de, ja) to ensure correct accent and intonation. This applies to all 23 supported languages, including Arabic, Chinese, and Korean.

Any of the 8 voices can speak any supported language, allowing consistent character voices across different languages in your stream or application.

  • Supported: Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, Turkish.
  • Voices are language-agnostic; use Aria for French or Felix for German without changing voices.

Unsupported languages list

The following languages are not currently supported: Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, Bengali, and others not listed above.

Request Limits Explained

Sound wave illustration for: Request Limits Explained

Understanding request limits is crucial for maintaining consistent audio generation during live streams or batch processing. The API enforces a standard rate limit of 60 requests per minute per active API key. This cap ensures fair usage and stable server performance for all indie developers and streamers relying on the service.

60 requests per minute limit

Each API key is allowed a maximum of 60 requests per minute once the account is active. This limit applies to all POST /v1/audio/speech calls, regardless of text length or format chosen. Monitoring your usage dashboard helps you stay within this threshold to avoid temporary 429 errors.

10 requests per minute initially

New accounts or those with limited balance start with a lower threshold of 10 requests per minute. This initial cap remains in place until you complete your first top-up. Once a top-up is processed, the limit automatically increases to the standard 60 requests per minute, ensuring smoother initial testing.

Queueing under high load

Requests are processed sequentially, meaning your audio synthesis job may wait in a queue if the service is busy. During peak usage times, you might experience slight delays beyond the standard processing time. This queuing mechanism ensures that each request is handled thoroughly without dropping connections.

Audio Generation Speed

Sound wave illustration for: Audio Generation Speed

For most content, audio generation runs at roughly real-time speeds, meaning the output duration matches the spoken length. This efficiency makes it suitable for standard voiceover workflows without significant waiting periods.

Roughly real-time generation

Long texts are processed at a pace that mirrors natural speech, allowing you to receive audio files without disproportionate delays relative to the content length.

25 characters per second average

The system processes approximately 25 characters of text per second of generated audio. This consistent rate helps in estimating production time for long-form videos or stream segments accurately.

Short phrases take a few seconds

Unlike long texts, short phrases require a few seconds to generate due to fixed processing overhead. Requests are handled sequentially, so high load may introduce slight queueing delays for rapid-fire requests.

Comparison with Competitors

Sound wave illustration for: Comparison with Competitors

Unlike many providers that impose tiered usage limits or volume-based discounts, our pricing remains a flat $10 per million characters regardless of scale. This simplicity eliminates complex calculations for indie developers and streamers, ensuring consistent costs whether you generate short alerts or long-form voiceovers.

We accept only USDT on the TRON network, avoiding credit card processing fees and bank delays. Top-ups are processed automatically within minutes, with no monthly subscriptions or recurring billing cycles to manage. Your balance never expires, allowing you to fund your account once and use it over time without pressure.

Flat rate vs tiered pricing ['Our pricing model is straightforward: $10 per 1 million characters. There are no volume discounts or tiered structures. This flat rate applies equally to small indie projects and larger stream productions, ensuring predictable budgeting without hidden volume-based adjustments.', 'Unlike competitors who charge different rates based on usage volume, we maintain a single rate. This consistency helps developers calculate exact costs for character generation without needing to track complex tier thresholds or negotiate bulk deals.'] ['Flat rate: $10 per 1 million characters', 'No volume-based discounts or tiered pricing', 'Consistent cost for all usage levels']

USDT only vs credit cards ['We accept USDT on the TRON network (TRC-20) exclusively. This method avoids credit card processing fees and allows for faster, direct top-ups. Minimum top-up is $5, credited automatically within minutes.', 'Credit cards, PayPal, and other traditional payment methods are not supported. This approach reduces overhead costs and streamlines the payment process for users familiar with cryptocurrency transactions, particularly in the gaming and streaming communities.'] ['USDT on TRON network (TRC-20) only', 'No credit cards or PayPal', 'Minimum top-up: $5', 'Automatic credit within minutes']

No subscriptions vs monthly plans ['There are no monthly plans or subscriptions. You pay only for what you use, with no recurring charges. This flexibility is ideal for indie developers and streamers with variable content production schedules.', 'Balance does not expire, allowing you to top up once and use the credits over time. This model eliminates the need to manage recurring billing cycles and ensures you only pay for actual usage.', 'No minimum spend or seats required. This approach contrasts with competitors who often require monthly commitments or seat-based licensing, providing greater freedom for small teams and individual creators.'] ['No monthly plans or subscriptions', 'Pay only for what you use', 'Balance never expires', 'No minimum spend or seats required']

Legal & Compliance

Sound wave illustration for: Legal & Compliance

EchoWave voices are derived from licensed recordings in the LibriTTS-R dataset, which utilizes public-domain audiobook readings from LibriVox. This foundation ensures that the synthetic voices are built upon legally cleared source material, providing a clear lineage for your generated audio assets.

Licensed recordings from LibriVox

The core audio data originates from LibriVox, a platform dedicated to free public domain books. By leveraging these existing recordings, we avoid the need to recruit and pay individual voice actors for every project, keeping costs low for indie developers and streamers.

  • Source data comes from public domain audiobooks
  • Recordings are licensed under CC BY 4.0
  • No individual actor rights to clear

CC BY 4.0 source material

The underlying LibriTTS-R dataset is released under the Creative Commons Attribution 4.0 International license. This allows you to use the generated audio in your projects, provided you adhere to the license terms regarding attribution where applicable. This clarity simplifies legal compliance for commercial and non-commercial streams.

  • CC BY 4.0 licensed source data
  • Clear usage rights for generated audio
  • Simplified legal status for indie creators

Synthetic speech watermark

Every audio file generated by our API includes an inaudible digital watermark. This marks the content as synthetic speech, which can help with transparency and compliance in platforms that require disclosure of AI-generated media. The watermark does not affect audio quality or playback.

Developer Resources

Sound wave illustration for: Developer Resources

The API documentation provides a complete reference for the /v1/audio/speech endpoint, including request schemas, response formats, and error codes. It details the JSON body structure required for text-to-speech synthesis, covering all supported parameters like model, voice, speed, and response_format.

Code examples are provided in multiple languages to facilitate quick integration. Since the API follows the same request shape as the OpenAI speech endpoint, you can use official OpenAI SDKs by simply changing the base_url to https://api.ktvsuper.info/v1. Alternatively, use any HTTP client with the standard Authorization: Bearer header.

API documentation link

Access the full API reference to understand all available endpoints, including /v1/voices for listing voices and /v1/models for model details. The documentation includes ready-to-use cURL examples for immediate testing.

Code examples in multiple languages

Integrate using the OpenAI SDK or any standard HTTP library. Change the base_url to https://api.ktvsuper.info/v1 and provide your API key. Examples are provided for Python, JavaScript, and cURL.

Community support channels

For technical questions, refer to the documentation or contact support via email. There are no dedicated community forums or live chat channels; support is handled exclusively through email.

Frequently Asked Questions

Sound wave illustration for: Frequently Asked Questions

The pricing model is straightforward: you pay $10 per million characters, which equals $0.01 per 1,000 characters. This flat rate applies regardless of volume, voice choice, or audio format selected. Spaces and punctuation are counted as characters, so plan your input text accordingly. There are no monthly subscriptions, per-seat fees, or hidden processing charges.

How much does it cost to generate 1 minute of audio?

Audio generation speed is roughly real-time, averaging about 25 characters per second. Therefore, generating one minute of audio requires approximately 1,500 characters. At $0.01 per 1,000 characters, the cost is approximately $0.015 per minute of audio output.

Does the price change based on the voice selected?

No. Both the "tts-1" and "tts-1-hd" models use the same engine and pricing tier. All eight voice presets (Aria, Nova, Luna, Sage, Felix, Milo, Atlas, Orion) are included in the standard character rate. Switching voices does not incur additional fees.

Are there any fees for using different audio formats?

The API supports MP3, WAV, OPUS, FLAC, and PCM formats without extra charges. The cost is determined solely by the character count of the input text, regardless of the output format chosen. There are no per-format or per-voice extra costs.

How long does the free trial last?

New accounts created via Google sign-in receive $0.50 in free credit, valid for 7 days. This allows for approximately 50,000 characters of generation. Accounts created with an email and password start with zero credit and do not receive the trial bonus.

Can I top up with Bitcoin or Ethereum?

Top-ups are accepted exclusively via USDT on the TRON network (TRC-20). We do not support Bitcoin, Ethereum, PayPal, or credit cards. Transfers are credited automatically within minutes of confirmation. There are no monthly plans or subscription fees.

What is the minimum amount to top up?

The minimum top-up amount is $5. There is no maximum limit. You can add funds as needed to cover your usage. Balance does not expire, so you can top up at your convenience.

Do I lose my balance if I don't use it?

No. Your account balance never expires. You can top up once and use the credits over an extended period. There are no monthly usage requirements or expiration dates on your stored credit.

Are failed requests charged?

Failed requests are free. If synthesis fails for any reason, the charge is returned in full. You only pay for successfully generated audio files. This ensures you are not billed for errors or processing failures.

Can I use the API for commercial projects?

Yes. The API is designed for both indie developers and streamers. You can use the generated audio for commercial projects, including VTuber streams, video voiceovers, and game assets. The voices are based on licensed recordings from LibriTTS-R and LibriVox.

What happens if I exceed the rate limit?

The standard limit is 60 requests per minute per key. New accounts start with 10 requests per minute. If you exceed the limit, requests may queue and process sequentially. Under high load, generation may take longer than real-time speed.

Is there a limit to the length of text I can send?

Each request is limited to 4,096 characters. For longer texts, you must split the content into parts, generate audio for each part, and join the files. This ensures consistent quality and manageable processing times.

Can I change my API key if it gets compromised?

You can issue a new key at any time from your dashboard. The old key stops working immediately. This allows you to secure your account quickly without losing your balance or usage history.

Do you support SSML for better control?

No. The API does not support SSML, phoneme dictionaries, or custom pronunciation lexicons. Control is limited to voice selection, speed (0.25–4.0x), and language code. We do not offer emotion tags or detailed audio formatting options.

Is there a discount for high-volume users?

The pricing is flat: $10 per million characters regardless of volume. There are no tiered pricing structures or volume discounts. This consistency benefits indie developers who may have fluctuating usage patterns.

How do I know which voice to choose?

We offer eight presets: female voices (Aria, Nova, Luna, Sage) and male voices (Felix, Milo, Atlas, Orion). Each has a distinct tone, from warm narration to energetic promo. Test voices using the free trial to find the best fit for your content type.

Can I use the audio files offline?

Yes. Once generated, the audio file is returned in the response. You can download and store it locally or use it offline. The generated audio carries an inaudible watermark to mark it as synthetic speech.

Do you offer a refund if I am unsatisfied?

We do not offer manual refunds for satisfaction. However, failed requests are automatically credited back. If you generate audio that does not meet your needs, the cost is still incurred. There is no money-back guarantee for successful generations.

Is my data secure when I send text to the API?

Input text is used only to produce the audio and is not used for training. Generated audio is returned in the response and not stored. Request metadata (time, characters, status) is kept for billing purposes only.

Can I integrate this API with my WordPress site?

There is no official WordPress plugin. You can integrate via plain HTTP or the OpenAI SDK by changing the base URL. Since we do not provide a custom SDK or browser extension, integration requires standard web development skills.

Is there a mobile app for managing my account?

No. There is no dedicated mobile app. You can manage your account, view balance, and generate keys via any web browser on any device. The API itself is accessible from mobile applications via HTTP requests.

Listen to the 8 voices

Each sample below was generated by this API from the same sentence. Each request is processed individually; under load, requests may queue, taking a few seconds for short phrases or roughly real-time for longer texts.

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. openai-python on GitHub
  4. W3C WAI: making audio and video accessible
  5. Node.js documentation
  6. openai-node on GitHub
  7. ISO 639 language codes
  8. IETF RFC 7845: Ogg Opus
  9. FFmpeg documentation
  10. OpenAI API reference: create speech
  11. Python documentation
  12. LibriTTS-R corpus (OpenSLR 141)

More on this site