Frequently asked questions
Which voice is best for Arabic content?
All eight voices support Arabic. You can select any preset, such as Aria or Felix, and set the language parameter to "ar". The API does not assign specific voices to specific languages, allowing you to choose the tone that fits your character’s personality regardless of the language spoken.
Is this API better than ElevenLabs?
Comparison depends on your needs. We offer fixed pricing at $10 per million characters with no monthly plans. ElevenLabs may offer different voice models and features like voice cloning, which we do not provide. Our API uses standard OpenAI-compatible endpoints, making integration straightforward for developers already familiar with that ecosystem.
How does the price compare to others?
Our rate is $10 per million characters, or $0.01 per 1,000 characters. This is a flat fee with no tiered pricing or subscriptions. Failed requests are free. We do not charge extra for different audio formats or voice models, keeping costs predictable for indie developers and streamers managing tight budgets.
Can I use this for YouTube videos?
Yes. The generated audio files are synthetic speech suitable for content creation. Each file includes an inaudible watermark marking it as synthetic. You can download the audio via the API response and upload it directly to YouTube. There are no restrictions on commercial use or content type for your videos.
Does it support SSML?
No, this API does not support SSML. You send plain text input, and the engine synthesizes it using the selected voice and speed parameters. If you need precise control over pauses or pronunciation, you must manage that logic in your application code before sending the request to the API.
How many voices are available?
There are eight voice presets available. Four female voices include Aria, Nova, Luna, and Sage. Four male voices include Felix, Milo, Atlas, and Orion. Each voice has a distinct tone, ranging from warm and clear to deep and documentary-style, allowing you to match specific character archetypes for your VTuber projects.
Is there a limit on usage?
You are limited to 60 requests per minute per API key. Each request can process up to 4,096 characters. If you exceed this limit, requests may queue or fail. Longer texts must be split into chunks by your application. Failed requests do not count against your character balance or incur charges.
Can I clone my own voice?
No, custom voice cloning is not offered. You cannot upload your own voice samples to create a unique voice model. You must choose from the eight existing presets. These voices are based on licensed recordings from LibriTTS-R and LibriVox, ensuring clear, licensed synthetic speech without the need for complex voice training processes.
What is the latency like?
Typical speed is roughly real-time, generating about 25 characters per second of audio. Short phrases may take a few seconds to process. Requests are handled sequentially, so high load can cause queuing delays. We do not offer streaming or WebSocket endpoints for real-time partial audio delivery during synthesis.
Do I need to pay for each format?
No, the price is consistent regardless of the audio format. You can choose between mp3, wav, opus, flac, or pcm without any additional cost. The rate remains $10 per million characters for all supported formats. This simplifies billing for developers who need different formats for various streaming or playback scenarios.
Is the API reliable?
We do not provide an SLA or uptime guarantee. The API processes requests sequentially, which may lead to delays during peak usage. Failed requests are free, ensuring you are not charged for errors. For critical production use, consider implementing retry logic in your application to handle potential transient network or server issues.
Can I integrate with OBS?
Yes, by using our API to generate audio files. You can create a script that calls the API endpoint and saves the output to a file or pipe. OBS can then play this audio via a media source or input device. Since we do not offer an OBS plugin, you must handle the integration via your own script.
What happens if I run out of credit?
You can top up your balance using USDT on the TRON network (TRC-20). Minimum top-up is $5, and credits are added automatically within minutes. There are no monthly plans or subscriptions, so you only pay for what you use. Your balance does not expire, allowing you to top up whenever needed.
How do I contact support?
Support is available via email only. There is no live chat or phone number. You can find the support email address on our contact page. Response times are not guaranteed, so plan your development or streaming schedules accordingly. We do not offer priority support tiers or dedicated account managers.
Evaluation Criteria
For VTubers and indie developers, evaluating a TTS API requires balancing voice distinctiveness with technical reliability. The selection process prioritizes audio fidelity, cost structure, and integration complexity to ensure seamless character voice generation during live streams.
Voice quality and variety
The API offers eight distinct voice presets, including female options like Aria and Luna, and male choices such as Felix and Atlas. Each voice is grounded in licensed recordings from LibriTTS-R, ensuring consistent quality across 23 supported languages.
- 8 unique voice presets with distinct tonal characteristics
- Support for 23 languages with automatic language detection
- Inaudible synthetic watermark in all generated audio files
Pricing transparency
Costs are straightforward at $10 per million characters, with spaces and punctuation counting toward the total. Failed requests incur no charge, and the billing model lacks hidden fees, subscriptions, or monthly commitments, making it predictable for variable usage.
- $0.01 per 1,000 characters
- Free trial credit of $0.50 for new Google sign-ins
- No charges for failed synthesis requests
Ease of integration
The endpoint mirrors the OpenAI speech API structure, allowing developers to use existing OpenAI SDKs by simply adjusting the base URL. This compatibility reduces development time, as standard JSON requests with model, input, and voice parameters work out of the box.
- Compatible with official OpenAI SDKs
- Standard JSON request body format
- Simple API key authentication via Bearer token
Voice Catalog Analysis
EchoWave provides eight distinct voice presets, each designed for specific narrative roles. The catalog includes four female voices—Aria, Nova, Luna, and Sage—and four male options: Felix, Milo, Atlas, and Orion. This variety allows VTubers to match character personalities precisely, from energetic promos to calm documentary styles.
8 distinct voice presets
The API offers eight fixed voice models rather than infinite customization. Users select from warm narrators, bright upbeat speakers, soft storytellers, or deep documentary voices. Each preset is based on licensed recordings from LibriTTS-R, ensuring consistent audio quality across all characters.
- Aria: warm, clear narrator
- Nova: bright, upbeat
- Luna: soft, storytelling
- Sage: calm, instructional
- Felix: energetic, promo
- Milo: friendly, conversational
Arabic language support
All eight voices support Arabic, alongside 22 other languages including English, Chinese, and Japanese. The system guesses the language from the script, but setting the language parameter ensures accuracy for non-English Latin text. This makes it viable for multilingual streamers and global VTuber audiences.
Voice characteristics overview
Voices are synthetic, marked by an inaudible watermark. They lack emotion tags or SSML, but offer clear, natural diction suitable for character voice generation. The fixed nature of the voices ensures consistent branding, though users cannot upload custom audio or clone specific voices.
Pricing Comparison
EchoWave charges $10 per million characters, making it a cost-effective choice for streamers generating daily content. This transparent pricing model ensures predictable costs for VTuber voice generation without hidden fees.
$10 per million characters
The rate is $0.01 per 1,000 characters, including spaces and punctuation. Failed requests are free, ensuring you only pay for successful synthesis.
Free trial availability
New accounts signing in with Google receive $0.50 of credit, valid for 7 days. Email-based accounts start without trial credit.
USDT payment option
Top-ups use USDT on the TRON network (TRC-20), with a $5 minimum. Credits appear automatically within minutes of transfer.
- No cards or PayPal accepted
- Balance does not expire
- Minimum top-up: $5
Technical Features
The API endpoint mirrors the OpenAI speech structure, allowing indie developers to reuse existing codebases by simply adjusting the base_url to https://api.ktvsuper.info/v1. This compatibility ensures that standard OpenAI SDKs function correctly with minimal configuration changes, streamlining the integration process for streamers and VTuber developers who already utilize these libraries.
OpenAI SDK compatibility
By changing the base URL to https://api.ktvsuper.info/v1, you can use official OpenAI SDKs directly. This reduces boilerplate code and leverages familiar authentication patterns using your unique API key.
- Same request shape as OpenAI speech endpoint
- Works with official OpenAI SDKs
- Simple
base_urlconfiguration change
Multiple audio formats
Generate audio in mp3, wav, opus, flac, or pcm formats. MP3 is suitable for general streaming due to its balance of quality and file size, while WAV and FLAC offer lossless quality for high-fidelity VTuber recordings. PCM is available for raw data processing pipelines.
- mp3, wav, opus, flac, pcm
- MP3 best for general streaming
- WAV/FLAC for lossless quality
Rate limits and speed
The API allows 60 requests per minute per key (10 per minute before the first top-up). Generation speed is roughly real-time (25 characters per second), though short phrases may take a few seconds. Requests are processed sequentially, so heavy loads may introduce queue delays.
- 60 requests per minute per key
- Roughly real-time generation speed
- Sequential processing may cause queues
Use Cases for Streamers
Streamers and VTuber developers require reliable, low-latency voice generation to maintain engagement without breaking immersion. This API provides distinct character voices and predictable pricing, making it ideal for automated content pipelines where consistency and cost-efficiency matter more than human nuance.
VTuber character voices
Use the eight preset voices to assign distinct personas to your VTuber models. For example, use 'aria' for a calm narrator or 'felix' for energetic promos. Since voices are based on licensed recordings, you get consistent, high-quality synthetic speech that matches your character’s visual style without needing custom voice cloning.
- Assign 'luna' for soft storytelling segments
- Use 'atlas' for neutral news-style updates
- Apply 'orion' for deep documentary tones
Chat response automation
Automate responses to viewer chat by sending text to the TTS endpoint. With a rate limit of 60 requests per minute, you can handle moderate chat volumes efficiently. Split longer messages into 4096-character chunks to ensure synthesis completes without errors, keeping your stream interactive without manual intervention.
- Process up to 60 requests per minute
- Split text to stay under 4096 chars
- Use USDT top-ups for sustained usage
Video content voiceovers
Generate voiceovers for pre-recorded videos or live streams using formats like MP3 or WAV. The API processes requests sequentially, so plan your queue accordingly for longer texts. Since generated audio is not stored, you retrieve the file directly in the response, keeping your workflow simple and secure.
- Supports MP3, WAV, Opus, FLAC, PCM
- Real-time speed for long texts
- No storage of generated audio files
Pros and Cons
EchoWave delivers value for indie VTubers and streamers, primarily due to its transparent and low-cost pricing model. At $0.01 per 1,000 characters, it offers a competitive rate compared to many competitors while maintaining high audio clarity through licensed LibriTTS-R recordings. Integration is straightforward, leveraging the familiar OpenAI SDK structure, which allows developers to swap base URLs without rewriting logic. This ease of use makes it suitable for rapid prototyping and consistent character voice generation across long streams.
Pros: Low cost, easy integration
The pricing structure is simple: $10 per million characters with no hidden fees or subscription tiers. Failed requests are refunded instantly, ensuring you only pay for successful synthesis. The API mimics the OpenAI speech endpoint, meaning existing OpenAI SDKs work by simply adjusting the base URL. This compatibility reduces development time and lowers the barrier to entry for indie developers who need reliable TTS without complex custom libraries.
- Transparent per-character billing ($0.01/1k chars)
- OpenAI SDK compatibility for quick setup
- Instant refund for failed requests
Cons: Fixed voices, no streaming
EchoWave offers eight fixed presets (Aria, Nova, Luna, Sage, Felix, Milo, Atlas, Orion) rather than dynamic voice cloning or custom design. These voices are based on licensed recordings and cannot be modified for nuance or emotion. Additionally, the API does not support streaming or WebSocket connections; audio is returned as a complete file after synthesis. This means you cannot hear partial audio in real-time, which may affect live interactive scenarios where immediate feedback is critical.
- Eight static voice presets only
- No custom voice cloning available
- No real-time streaming or WebSocket support
- Audio returned as complete files only
Limitations: Rate limits, email support
The API enforces a rate limit of 60 requests per minute per key, with a lower initial limit of 10 requests per minute before the first top-up. Under heavy load, requests may queue, potentially causing delays in generating audio for long texts. Support is exclusively via email, with no live chat, phone, or SLA guarantees. This lack of direct support channels may slow issue resolution for streamers needing immediate assistance during peak hours.
- 60 requests/minute limit (10 initially)
- Requests may queue under load
- Email support only; no live chat
- No SLA or uptime guarantee
Integration Guide
Setting up your API key is the first step to integrating EchoWave into your VTuber pipeline. Navigate to the /get-api-key/ page to sign in via Google or email. Your unique key appears immediately, ready for use in your application.
Keep your key secure, as it controls your billing. You can generate a new key at any time, which instantly invalidates the previous one, ensuring tight security for your streaming operations.
Setting up API keys
Navigate to the /get-api-key/ page to sign in via Google or email. Your unique key appears immediately, ready for use in your application.
Keep your key secure, as it controls your billing. You can generate a new key at any time, which instantly invalidates the previous one, ensuring tight security for your streaming operations.
- Sign up with Google or email
- Receive key immediately on dashboard
- Issue new key to revoke old one
Making first request
Use the official OpenAI SDK by changing the base_url to https://api.ktvsuper.info/v1. Send a POST request to /v1/audio/speech with your JSON body containing the text, voice, and model.
Ensure your text is under 4096 characters. The API returns an audio file directly in the response, which you can save or stream to your broadcasting software.
- Set base_url to https://api.ktvsuper.info/v1
- Use POST /v1/audio/speech endpoint
- Include Authorization: Bearer tts-... header
Handling errors
If synthesis fails, the charge is returned in full, so you can retry without penalty. Check your rate limits (60 requests per minute) if you receive 429 errors.
For persistent issues, contact support via email. There is no live chat or phone support, so allow time for a written response to your technical query.
- Failed requests are free
- Rate limit: 60 req/min
- Support via email only

Audio Quality Assessment
Audio fidelity varies by format, with WAV and FLAC offering lossless quality suitable for archival, while MP3 and OPUS provide efficient compression for live streaming. PCM delivers raw, uncompressed data for immediate processing.
Every generated file contains an inaudible watermark marking it as synthetic speech, ensuring transparency without affecting audio clarity. This watermark persists across all supported formats, including WAV, MP3, OPUS, FLAC, and PCM.
Synthetic watermark presence
The API embeds an inaudible watermark in every generated audio file, clearly marking it as synthetic speech. This feature ensures transparency for VTuber streams and video content without impacting the listening experience or audio fidelity.
Clarity and naturalness
Voices like Aria and Luna offer warm, clear narration, while Felix provides energetic delivery. Clarity remains consistent across all 23 supported languages, though non-English Latin text requires explicit language setting to avoid English default guessing. Generation speed averages roughly real-time for long texts, with short phrases taking a few seconds.
Naturalness depends on the preset chosen; Sage offers calm instructional tones, while Orion provides deep documentary styles. The audio quality is consistent, derived from licensed LibriTTS-R recordings, ensuring a uniform synthetic character across all outputs.
- Voice presets include Aria, Nova, Luna, Sage, Felix, Milo, Atlas, and Orion.
- Supports 23 languages with accurate pronunciation when language code is specified.
- Generation speed is approximately 25 characters per second for long texts.
Format quality differences
MP3 and OPUS are optimal for streaming due to lower bandwidth usage, while WAV and FLAC provide lossless quality for high-fidelity needs. PCM offers raw audio data for immediate processing without compression artifacts.
Format selection does not affect the synthetic watermark or voice characteristics. Choose based on your storage and bandwidth constraints, ensuring the quality meets your VTuber stream or video content standards.
- Supported formats: MP3, WAV, OPUS, FLAC, PCM.
- WAV and FLAC offer lossless quality for archival.
- MP3 and OPUS are efficient for live streaming.
Customer Support Experience
EchoWave relies exclusively on email for all customer inquiries, eliminating the expectation of immediate, real-time assistance. This model suits developers who prefer asynchronous communication and can document technical issues with precision rather than seeking instant verbal clarification during a live stream.
There is no live chat widget, toll-free number, or dedicated account manager for standard accounts. Support is entirely ticket-based, which keeps operational costs low and allows us to pass those savings directly to you as lower API rates.
Email support only
All questions, from billing disputes to technical integration queries, are handled via a single support email address. You can reference your inquiry for tracking, ensuring that your specific context is preserved without the need to re-explain details in a transient chat window.
- Single point of contact for all issues
- Ticket ID for tracking conversation history
- Written record of all support interactions
No live chat or phone
Unlike many consumer-focused TTS providers, EchoWave does not offer phone lines or real-time chat support. This absence is intentional, designed to reduce overhead and maintain a lean support structure that prioritizes depth of technical answer over speed of initial response.
Response time expectations
While specific response times are not guaranteed by an SLA, replies typically arrive within 24 hours. For critical billing issues, such as confirming a USDT top-up, responses are often faster, but high-volume periods may cause slight delays.
- Standard queries: ~24 hours
- Billing issues: Often faster
- No SLA guarantee
Security and Privacy
EchoWave prioritizes privacy by using input text solely for audio synthesis without retaining the content for training purposes. Request metadata, such as timestamps and character counts, is stored exclusively for billing accuracy.
Data usage policies
Generated audio files are returned in the response and are not stored on our servers. This ensures that your specific voice outputs remain transient and private.
No training on input text
Unlike some competitors, we do not use your input text or generated audio to train our models. Your data remains yours, ensuring strict separation between your content and our engine improvements.
API key security
Each account holds one active key, which can be revoked instantly. Issuing a new key immediately invalidates the old one, providing simple yet effective access control without complex permission layers.
- Instant key invalidation
- One active key per account
- Bearer token authentication
Final Verdict
EchoWave provides a straightforward, reliable TTS API tailored for developers who need consistent character voices without complex infrastructure. The fixed voice catalog simplifies asset management, while the transparent per-character pricing eliminates billing surprises. This clarity makes it a practical tool for small-scale projects and indie games.
Suitable for indie developers
The API’s OpenAI SDK compatibility allows indie developers to integrate TTS with minimal code changes. With a simple base URL adjustment and standard JSON requests, you can add voice synthesis to your projects quickly. The lack of custom cloning is a trade-off, but the eight distinct presets cover most character archetypes effectively.
- Easy integration with existing OpenAI SDKs
- No complex infrastructure required
- Clear API documentation
Ideal for VTuber content
For VTubers, consistent character voice is paramount. EchoWave’s eight presets offer distinct personalities, from the warm narration of Aria to the deep documentary tone of Orion. The synthetic watermark ensures clear identification of AI-generated speech, which is useful for content creators managing copyright and authenticity expectations.
- Distinct voice personalities for diverse characters
- Synthetic watermark for authenticity
- Support for multiple languages including Japanese and English
Cost-effective solution
At $10 per million characters, EchoWave is highly affordable for budget-conscious creators. The free trial provides 50,000 characters, allowing you to test voice quality and integration before committing. With USDT payments and no monthly subscriptions, you only pay for what you use, making it ideal for variable content production cycles.
- Pay-per-character model
- Free trial with 50,000 characters
- No monthly subscriptions
Recommendations
Start by creating an account to access the $0.50 free trial, which provides approximately 50,000 characters of credit valid for seven days. This allows indie developers to test voice quality and integration without immediate financial commitment, ensuring the synthetic voices meet their VTuber character requirements before scaling.
Start with the free trial
The trial credit is automatically applied upon signing in with Google, making it easy to begin testing immediately. Use this period to evaluate how well each of the eight voice presets suits your specific content style, from the warm narration of Aria to the deep documentary tone of Orion.
- Google sign-in grants instant $0.50 credit
- Valid for 7 days
- Approximately 50,000 characters
Use MP3 for general streams
For most streaming and video use cases, the MP3 format offers the best balance of file size and audio fidelity. It is widely compatible with OBS, VLC, and most streaming software, ensuring smooth playback without significant CPU overhead compared to uncompressed WAV or FLAC files.
- MP3: Best balance of size and quality
- WAV/FLAC: Higher quality, larger files
- Opus/PCM: Specific codec needs
Monitor rate limits closely
Keep a close watch on your usage, as the API enforces a limit of 60 requests per minute. If you are generating long-form content, split your text into chunks under 4,096 characters to ensure efficient processing. Remember that requests are queued sequentially, so heavy load may increase latency.
- 60 requests per minute limit
- 4,096 characters max per request
- Requests processed sequentially
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.
23 languages
Any voice speaks any of these languages.
Sources and further reading
- Wikipedia
- openai-python on GitHub
- Node.js documentation
- openai-node on GitHub
- LibriTTS-R corpus (OpenSLR 141)
- OpenAI API reference: create speech
- LibriVox public-domain audiobooks
- W3C WAI: making audio and video accessible
- Tether (USDT)
- Creative Commons Attribution 4.0
- ISO 639 language codes
- IETF RFC 7845: Ogg Opus
