Frequently asked questions
What is a text-to-speech API?
It is a cloud API where you send text via POST and receive an audio file. The endpoint is https://api.3178.info/v1/audio/speech. It uses the same request shape as the OpenAI speech endpoint, allowing you to use official OpenAI SDKs by simply changing the base_url to https://api.3178.info/v1 and providing your API key.
How does French pronunciation work in TTS?
French is one of our 23 supported languages. Any voice can speak any supported language. If you do not set the language field, the system guesses it from the script, but for non-English Latin-script text like French, you should explicitly set the language parameter to ensure precise pronunciation.
Which voices are best for museum guides?
For museum guides, we recommend aria (warm, clear narrator), luna (soft, storytelling), or atlas (neutral, news). These voices provide consistent volume and clear narration suitable for audioguides. All eight voices are licensed recordings based on LibriTTS-R and LibriVox public-domain readings, ensuring high quality for tourism content.
Can I generate audio in multiple languages?
Yes, we support 23 languages including French, German, Spanish, and Italian. Any voice can speak any of these languages. You can specify the language using an ISO code in the request. If omitted, the system guesses the language from the script, defaulting to English for Latin text.
What is the character limit per request?
Each request accepts up to 4096 characters of text in the input field. If your narration text exceeds this limit, you must split it into parts of up to 4096 characters and join the resulting audio files. This ensures efficient processing and adherence to the API's structural constraints.
How are characters counted in the API?
Characters include spaces and punctuation. The pricing is $10 per 1 million characters, or $0.01 per 1,000 characters. Every character in your input text contributes to the count, so ensure your text is concise to manage costs effectively while maintaining the quality of your audioguide narration.
Is the audio watermark audible?
No, every generated file carries an inaudible watermark that marks it as synthetic speech. This watermark is not audible to listeners but serves to identify the audio as AI-generated. This applies to all outputs regardless of the voice, language, or format selected for your audioguide project.
Do I need to split long texts manually?
Yes, you must split texts longer than 4096 characters into parts and join the audio files yourself. The API does not handle automatic splitting or joining. You should implement this logic in your application to ensure seamless playback of longer narrations for museum exhibits or tourism content.
What formats are supported for audio output?
We support mp3, wav, opus, flac, and pcm via the response_format field. Choose mp3 for efficient storage and web compatibility, or wav/flac for high-fidelity audioguide playback. The format selection allows you to balance audio quality with file size requirements for your specific tourism or museum application.
How does the speed parameter affect narration?
The speed parameter ranges from 0.25 to 4.0, allowing you to adjust the narration pace. This is useful for adapting to different listener needs or specific exhibit contexts. Adjusting speed does not change the character count or pricing, but it affects the duration of the generated audio file.
Are failed requests charged?
No, failed requests are free. If synthesis fails, the charge is returned in full. This ensures you only pay for successful audio generation. You can test your integration confidently, knowing that errors in processing or temporary issues will not incur costs against your account balance.
What is the free trial credit?
You receive $0.50 of credit (about 50,000 characters) valid for 7 days when signing in with Google. Accounts created with email and password start without trial credit. This allows you to test the API's quality and integration before committing to a top-up for your audioguide production.
How do I top up my account?
Top up using USDT on the TRON network (TRC-20) only. The minimum amount is $5, and credits are automatically added within minutes after the transfer. We do not accept cards, PayPal, or subscriptions. Your balance does not expire, allowing you to manage costs for long-term tourism projects.
First request
curl
curl https://api.3178.info/v1/audio/speech \
-H "Authorization: Bearer $TTS_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"tts-1","voice":"aria","input":"Hello from the API.","response_format":"mp3"}' \
--output hello.mp3Optimized for Audio Content
AudioGuides API delivers consistent narration and reliable volume normalization, ensuring professional audio for museum exhibits and tourism content. Our engine generates complete audio files via JSON responses, eliminating the variable latency of partial audio chunks. This allows your application to buffer and play audio instantly once synthesis is complete, providing a smooth listener experience without streaming delays.
Consistent Narration
Our TTS engine maintains steady pacing and tone across long texts, ideal for audioguides. Texts are split into 4096-character segments and joined seamlessly, preserving narrative flow. Voices like Aria or Atlas provide clear, neutral delivery suitable for historical or educational content.
- Uniform pacing across all segments
- Seamless audio joining
- Neutral delivery for historical content
Volume Normalization
Each generated file undergoes consistent volume processing, preventing sudden loudness shifts between phrases. This ensures a comfortable listening experience in varying acoustic environments, from quiet museum halls to noisy outdoor tours.
- No sudden volume spikes
- Comfortable listening experience
- Suitable for various environments
No Streaming Latency
Unlike streaming solutions, we return complete audio files as binary data in the response body. This eliminates the variable latency of partial audio chunks, allowing your app to buffer and play audio instantly once the synthesis is complete.
- Instant playback after synthesis
- No partial audio chunks
- Buffer-friendly design
Voice Selection for Exhibits
Selecting the right voice is critical for maintaining engagement in audioguide and museum environments. Our API offers eight distinct presets, each tuned for specific narration styles to match the tone of your exhibits.
Female Narrators
Aria provides a warm, clear narrator voice ideal for general exhibits. Nova offers a bright, upbeat tone suitable for interactive displays. Luna delivers a soft, storytelling voice for historical narratives, while Sage is optimized for calm, instructional content.
- Aria: warm, clear narrator
- Nova: bright, upbeat
- Luna: soft, storytelling
- Sage: calm, instructional
Male Narrators
Felix brings an energetic tone perfect for promotional audio or fast-paced tours. Milo provides a friendly, conversational style for user guides. Atlas delivers a neutral, news-style voice for factual information, and Orion offers a deep, documentary tone for serious exhibits.
- Felix: energetic, promo
- Milo: friendly, conversational
- Atlas: neutral, news
- Orion: deep, documentary
Tone Matching
Each voice is based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain audiobooks, ensuring consistent quality. You can pair specific voices with content types to maintain a cohesive audio identity across your entire guide system.
Multilingual Support
Our engine supports 23 languages, allowing you to deploy audio guides across diverse international markets without managing multiple voice models. Any voice can speak any of these languages, ensuring consistent quality regardless of the target audience's native tongue.
If the language is not set, it is guessed from the script (Latin text defaults to English). This reduces boilerplate in your integration while maintaining accuracy for standard use cases.
23 Languages
We support Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi, Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, and Turkish. This coverage handles the majority of global tourism and museum audiences.
- Arabic, Chinese, Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Hindi
- Italian, Japanese, Korean, Malay, Norwegian, Polish, Portuguese, Russian, Spanish, Swahili, Swedish, Turkish
Automatic Detection
If the language field is omitted, the system infers the language from the script. Latin script text defaults to English, so you must explicitly set the language for other Latin-based languages like French or Spanish to ensure correct pronunciation.
Explicit Language Codes
For non-English Latin-script text, set the language parameter to the ISO code. This guarantees precise pronunciation and avoids misinterpretation. Languages like Vietnamese, Indonesian, Thai, Ukrainian, Romanian, Czech, and Bengali are not supported.
Technical Implementation
The API endpoint at https://api.3178.info/v1 accepts standard JSON payloads. Authentication uses a Bearer token in the Authorization header. The request structure mirrors the OpenAI speech endpoint, allowing you to swap the base URL in existing OpenAI SDKs without code changes.
OpenAI Compatible
Send a POST request to /v1/audio/speech with model, input, voice, and format fields. The engine processes text synchronously, returning the audio file directly in the response body. No WebSocket or streaming connections are required.
JSON Responses
All metadata, including character count and request status, is returned in JSON format. This ensures easy parsing in backend services and mobile apps. Failed requests return error codes with clear messages for debugging.
Multiple Formats
Specify the output format using the response_format field. Supported formats include mp3, wav, opus, flac, and pcm. Choose mp3 for web delivery or wav for high-fidelity museum audio players.
Pricing Model
AudioGuides API charges $0.01 per 1,000 characters, including spaces and punctuation. Failed requests are refunded in full, ensuring you pay only for successful synthesis.
Per Character
Usage is calculated based on the exact character count of your input text. At $10 per 1 million characters, costs scale linearly with volume, making it ideal for large museum archives or tourism apps with extensive content libraries.
No Monthly Fees
There are no subscriptions, seats, or minimum spend requirements. You top up your balance using USDT (TRC-20) on the TRON network, with a $5 minimum deposit. Credits never expire, and unused funds remain available for future requests.
Transparent Costs
Your dashboard displays real-time balance, character usage, and remaining trial credit. Since billing is tied directly to request metadata, you can audit every dollar spent against specific audio generations without hidden tier restrictions or volume discounts.

Content Preparation
Prepare input text for synthesis by respecting character constraints and ensuring accurate length calculations. Each request supports up to 4096 characters, including spaces and punctuation, which directly impacts billing and processing time.
Text Length Limits
Limit each API call to 4096 characters. Exceeding this limit causes the request to fail. This constraint applies to the raw input string sent in the input field.
Character Counting
Count every character in the input text, including spaces, punctuation marks, and special symbols. For example, a comma or a period counts as one character. This total determines the cost at $0.01 per 1,000 characters.
- Include all whitespace and punctuation in your count.
- Spaces between words are counted as characters.
Splitting Logic
For longer audioguide scripts, split the text into chunks of 4096 characters or fewer. Generate audio for each chunk separately, then join the resulting audio files to create the final continuous narration. This ensures reliable synthesis and avoids request failures.
Audio Quality
Every generated audio file carries an inaudible watermark that marks it as synthetic speech. This metadata is embedded directly into the audio stream without affecting playback quality or volume levels, ensuring your audioguide content remains distinct from human-recorded tracks.
Watermarked Audio
The watermark is inaudible to listeners but detectable by software, providing a clear distinction between AI-generated narration and human voiceovers. This allows you to maintain transparency with your audience while preserving the professional audio quality expected in museum exhibits.
Synthetic Detection
Because the audio is synthetic, it can be reliably identified by AI detection tools. This is important for compliance in regions where labeled synthetic media is required, especially for large-scale tourism deployments or educational content distributed online.
Format Options
We support mp3, wav, opus, flac, and pcm formats to suit different playback systems. MP3 is recommended for mobile tourism apps due to its balance of quality and file size, while WAV or FLAC is ideal for high-fidelity museum kiosks requiring uncompressed audio.
- MP3: Recommended for mobile apps and web playback
- WAV/FLAC: Lossless quality for high-end kiosks
- OPUS: Efficient compression for streaming
- PCM: Raw audio for custom processing pipelines
Use Cases
AudioGuides API delivers reliable, multilingual narration for museum exhibits and tourism applications. Our engine ensures consistent volume and precise pronunciation across 23 languages, critical for professional audio experiences.
Museum Guides
Generate consistent voiceovers for static exhibits or interactive kiosks. The API handles text up to 4096 characters, allowing seamless integration with existing content management systems without complex streaming infrastructure.
- Consistent tone across all exhibit audio
- No streaming latency for kiosk playback
- Supports 23 languages for international visitors
Tourism Apps
Integrate localized audio into mobile tourism apps. Users receive instant audio responses for points of interest, with volume normalization ensuring clear playback in noisy outdoor environments.
- Instant audio generation for POIs
- Volume normalization for outdoor use
- Works with standard OpenAI SDKs
Educational Content
Produce audio for language learning or historical education. The API supports clear, neutral narration styles suitable for instructional content, with accurate pronunciation across multiple dialects.
- Clear narration for learning materials
- Accurate pronunciation across supported languages
- Easy integration with LMS platforms
Developer Experience
The API follows a familiar request shape, making integration straightforward for engineering teams already using standard HTTP clients. You send a POST request with JSON parameters and receive the audio file directly, requiring minimal custom code to integrate into your existing audio pipelines.
Our documentation provides ready-to-use examples, so you can start testing immediately. The consistent structure means you only need to configure the base URL and authentication header to begin generating speech for your audioguide content.
Simple API
The endpoint POST /v1/audio/speech accepts standard JSON bodies, mirroring the OpenAI speech format. This reduces the learning curve for developers familiar with modern TTS services. You simply provide the text, voice, and format, and the API returns the audio file.
Authentication is handled via a Bearer token in the Authorization header. A single active key per account simplifies credential management in your application's environment variables.
- Standard JSON request body
- Bearer token authentication
- Direct audio file response
SDK Support
While we do not provide a proprietary SDK, the API is compatible with the official OpenAI SDKs. You can use them by updating the base_url to https://api.3178.info/v1 and passing your API key.
This allows you to leverage existing client libraries in Python, Node.js, and other languages without writing custom HTTP parsing logic for audio responses.
Documentation
The /docs/ page details all parameters, including voice presets, supported formats, and rate limits. You can find specific examples for generating speech in French and other supported languages.
For quick testing, the /get-api-key/ page includes a ready-to-run cURL command. This allows you to verify connectivity and audio quality before integrating the API into your production environment.
Listen to the 8 voices
Each sample below was generated by this API from the same sentence. Selecting the right voice is critical for museum audioguides, where clarity and tone must match the exhibit's context.
23 languages
Any voice speaks any of these languages.
