Text-to-Speech API for Museum Audio Guides

By EchoGuide editorial team · Updated 2026-11-19

This page explains how EchoGuide’s text-to-speech API helps tourism and museum staff create natural-sounding audio guides without coding. Turn your guide scripts into ready-to-play MP3s with authentic German and international accents. We handle the technical details so you can focus on your exhibits.

You will find a clear breakdown of our 8 voice presets, 23 supported languages, and simple pricing. Learn how to generate audio via our API, manage long scripts, and understand your rights. Everything you need to launch an engaging audio experience for your visitors starts here.

Text-to-Speech API for Museum Audio Guides

Frequently asked questions

Is this API suitable for large museums?

Yes, the API handles high volume efficiently. You can send texts up to 4,096 characters per request and join audio files for longer scripts. With a rate limit of 60 requests per minute, it scales well for museum guide production workflows without requiring complex infrastructure.

How many languages does it support?

We support 23 languages, including German, English, French, Spanish, and Chinese. Any voice can speak any of these languages. If you do not set the language parameter, the system guesses it, but we recommend setting it explicitly for non-English Latin text to ensure accuracy.

Can I use the audio on my website?

Yes, you receive the audio file directly in the API response. You can store or play it wherever you like. Each generated file carries an inaudible watermark marking it as synthetic speech, but there are no restrictions on where you play the resulting MP3 or WAV files.

Do I need to credit the voices?

The voices are based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain readings. While attribution is good practice, the API output is a synthetic file. You are not required to credit specific voice actors, but you may mention the synthetic nature if desired.

Is there a limit on audio length?

Each API request accepts up to 4,096 characters. For longer guide scripts, split the text into parts and join the resulting audio files. This allows you to create unlimited-length audio guides by chaining multiple requests together seamlessly within your workflow.

Can I use the API for commercial tours?

Yes, there are no restrictions on commercial use. You pay only for characters used at $0.01 per 1,000 characters. There are no monthly plans, seats, or minimum spends, making it ideal for commercial tour operators who want predictable costs based on usage volume.

How accurate is the German accent?

Our voices are trained on high-quality datasets and support German natively. You can use voices like Aria or Nova for clear narration. Since any voice can speak any supported language, you can choose the tone that best fits your museum’s brand while maintaining accurate pronunciation.

Are there any restrictions on use?

The main limit is 4,096 characters per request and 60 requests per minute. Input text is used only for synthesis and not for training. Generated audio is not stored by us. There are no content restrictions, so you can generate guides for any topic or audience.

Can I download the voices permanently?

No, you download the generated audio files, not the voices themselves. The voices are presets like Aria, Nova, or Felix. Once generated, you own the audio file, but you cannot extract the voice model or use it independently outside of our API format.

Is the service reliable for peak hours?

Requests are processed sequentially, so under heavy load, a request may wait in a queue. Typical speed is roughly real-time. We do not offer an SLA or uptime guarantee, but the system is designed to handle standard tourism volumes efficiently without complex scaling needs.

Do I need a developer for integration?

No, our API follows the standard OpenAI request shape. You can use simple curl commands or no-code tools that support HTTP requests. The endpoint is straightforward: send text and receive audio. Many tourism staff integrate it using basic automation tools without writing code.

Can I change the voice after generating?

No, the voice is determined at generation time. To change the voice, you must send a new request with the desired voice preset (e.g., switching from Aria to Felix). The audio file itself does not contain editable voice parameters; it is a final audio file.

What happens if the API goes down?

We do not provide an SLA or uptime guarantee. If the service is unavailable, you cannot generate audio until it returns. Failed requests are free, so you are not charged for attempts made during downtime, but you may experience delays in producing your guide audio.

Is there a free tier for testing?

Yes, signing in with Google gives you $0.50 of credit (about 50,000 characters) valid for 7 days. Email accounts start without trial credit. This allows you to test voice quality and integration before purchasing USDT top-ups for your museum’s audio guide production.

Why use TTS for museums?

Sound wave illustration for: Why use TTS for museums?

Static audio guide files require physical swapping or complex reprogramming when exhibit texts change. With our API, you update the script in your system and generate fresh audio instantly, keeping your museum’s narrative current without technical delays.

Museum visits vary greatly in crowd size and language needs. TTS allows you to scale audio production for new languages or seasonal exhibits on demand, avoiding the high cost of hiring multiple voice actors or recording studios for every update.

Quick content updates

When an exhibit description changes or a temporary display is added, you simply send the new text to the API. The system generates the audio file quickly, allowing you to push updates to your guide devices or web app without waiting for studio recording schedules.

Cost-effective scaling

Generating audio costs $0.01 per 1,000 characters. This means you only pay for what you use, making it feasible to maintain audio for dozens of exhibits or multiple language versions without the fixed costs of recording studios or long-term contracts.

Consistent voice quality

Your visitors hear the same warm, clear narrator across all exhibits. With eight distinct voice presets, you can maintain a consistent brand voice for your entire museum, ensuring a professional and unified experience from the entrance to the final exhibit.

Our API advantages

Sound wave illustration for: Our API advantages

EchoGuide simplifies audio guide production by providing eight distinct voice presets and support for 23 languages, ensuring your museum content sounds natural and accessible to international tourists without requiring technical expertise.

8 unique voice presets

Choose from eight licensed voices based on LibriTTS-R data, ranging from warm narrators to energetic promoders. Each voice offers a distinct tone suitable for different exhibition styles, from soft storytelling to clear instructional guidance.

  • Aria: Warm, clear narrator
  • Felix: Energetic, promo style
  • Atlas: Neutral, news delivery
  • Luna: Soft, storytelling tone

23 languages supported

Our API handles 23 languages, including German, English, French, Spanish, and others crucial for tourism. You can specify a language code or let the system guess from Latin scripts, ensuring accurate pronunciation for diverse visitor groups.

Simple API integration

Send text via a standard POST request and receive an audio file. The endpoint mirrors the OpenAI speech format, allowing you to use existing OpenAI SDKs by simply adjusting the base URL to our API address.

  • POST request to /v1/audio/speech
  • JSON body with text and voice
  • Standard Authorization header with Bearer key

Voice variety comparison

Sound wave illustration for: Voice variety comparison

Choose from eight distinct voice presets tailored for museum and tourism contexts. Our library includes four male and four female voices, each designed to deliver clear, natural-sounding narration for audio guides.

Select the perfect tone for your specific exhibit or tour type, ensuring consistency across all your content.

Male voices: Felix, Milo, Atlas, Orion ['Felix offers an energetic, promo-style delivery ideal for dynamic city tours. Milo provides a friendly, conversational tone suitable for interactive exhibits. Atlas delivers a neutral, news-like clarity perfect for factual historical information. Orion uses a deep, documentary style for immersive storytelling.', {'h3': 'Female voices: Aria, Nova, Luna, Sage', 'paras': ['Aria provides a warm, clear narrator voice for general use. Nova offers a bright, upbeat tone for lively attractions. Luna creates a soft, storytelling atmosphere for historical narratives. Sage delivers a calm, instructional voice for detailed guide information.', 'bullets']}, {'h3': 'Each with distinct tone', 'paras': ['Each voice is based on licensed recordings from LibriTTS-R (CC BY 4.0) and LibriVox public-domain readings, ensuring authentic quality. You can match the voice to the content style, whether it is a quick promo or a deep historical account.', 'bullets']}] A minimalist illustration showing eight distinct voice waveform icons arranged in two rows, representing male and female tones, against a soft museum background with blurred exhibit shelves

Language support for tourists

Sound wave illustration for: Language support for tourists

Support your international visitors with clear, natural audio in their native tongue. We cover major tourist languages like German, English, French, and Spanish, ensuring your museum scripts sound professional and easy to understand for global audiences.

German, English, French, Spanish

These languages cover the majority of international tourist demographics. Each voice preset adapts naturally to the linguistic nuances of these regions, providing a consistent listening experience regardless of the visitor's origin.

Asian and European languages

Beyond the primary four, we support 19 additional languages including Japanese, Korean, Chinese, Italian, Portuguese, Russian, and Turkish. This allows you to serve diverse groups without needing multiple specialized voice talents or external translation services.

Auto-detection for Latin scripts

If you do not specify a language, the API guesses it from the script. Latin text defaults to English, but for non-English Latin-script content, you should explicitly set the language parameter to ensure accurate pronunciation.

Pricing simplicity

Sound wave illustration for: Pricing simplicity

EchoGuide’s pricing model is straightforward and transparent. You pay only for the text you actually convert, with a flat rate of $0.01 per 1,000 characters. This cost includes every space, comma, and punctuation mark, ensuring predictable expenses for your audio guide projects without hidden surprises.

Pay per character used

Each request costs $0.01 per 1,000 characters, including spaces and punctuation. If a synthesis fails, the charge is returned in full, so you only pay for successful audio generation.

No monthly fees

There are no subscriptions, seats, or minimum spend requirements. Whether you generate 100 characters or 100,000, the rate remains consistent. Top-ups are available via USDT (TRC-20) on the TRON network, with a $5 minimum, and your balance never expires.

Free trial available

Start testing immediately with $0.50 of credit, enough for about 50,000 characters. This trial is valid for 7 days and is automatically granted when you sign in with Google. Accounts created with email and password start without this trial credit.

Technical requirements

Sound wave illustration for: Technical requirements

Audio guides require reliable, consistent narration without the overhead of managing a full development team. Our API is designed to integrate directly into existing museum websites and mobile apps, allowing staff to update guide scripts instantly. You can trigger audio generation for new exhibits or corrected translations with minimal technical friction.

No coding needed for basics

Start by using ready-made curl examples or our web interface to generate audio files. The system handles text processing and audio synthesis, delivering ready-to-play audio files (MP3, WAV, etc.) that can be embedded directly into your guide app. For more complex workflows, the API accepts standard JSON requests, making it easy to connect with no-code tools like Zapier or Make.

  • Use provided curl examples for immediate testing
  • Embed generated MP3s directly into web pages
  • Connect via no-code automation platforms easily

API works with web tools

The endpoint accepts standard HTTP POST requests, ensuring compatibility with most modern web frameworks. You can send text up to 4,096 characters per request and receive audio in seconds. This simplicity allows museum staff to automate the creation of audio files for new exhibit labels without building custom infrastructure.

OpenAI SDK compatible

Because our 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.ace44.info/v1. This means developers can leverage familiar libraries and authentication patterns, reducing the learning curve for integrating TTS into existing tourism software stacks.

Audio format options

Sound wave illustration for: Audio format options

Choose the audio format that best fits your museum’s delivery method. MP3 offers broad compatibility for web players and standard devices, ensuring your guide scripts play smoothly for tourists without requiring specific codecs. This format balances quality and file size effectively for general use.

MP3 for broad compatibility

MP3 is the default choice for most tourism applications. It ensures your audio guides play on virtually any smartphone, tablet, or dedicated audio player found in visitor centers. The format provides sufficient clarity for spoken content while keeping file sizes manageable for quick downloads over varying network conditions.

  • Works on all standard devices
  • Optimized for web playback
  • Balanced quality and size

WAV for high fidelity

Select WAV when you need uncompressed, high-fidelity audio. This format is ideal for professional installations or archival purposes where audio integrity is paramount. It preserves the full nuance of the TTS engine’s voice presets without any lossy compression artifacts.

  • Uncompressed audio quality
  • Ideal for archival use
  • Preserves voice nuances

OPUS, FLAC, PCM support

For specialized needs, we also support OPUS, FLAC, and PCM. OPUS offers excellent compression for low-bandwidth scenarios. FLAC provides lossless compression for storage efficiency. PCM delivers raw audio data for immediate processing. Use these formats when your specific hardware or software pipeline requires them.

  • OPUS for low bandwidth
  • FLAC for lossless storage
  • PCM for raw data
Audio format options

Handling long scripts

Sound wave illustration for: Handling long scripts

Long museum scripts often exceed the 4,096-character limit per request. You can split your guide text into smaller parts and generate audio for each segment individually.

Once all segments are created, you can merge the resulting audio files to produce a single, seamless track. This ensures your entire tour guide is ready for playback without losing quality.

Split text into chunks

Divide your script into segments of up to 4,096 characters. Each part is processed as a separate API request, allowing you to handle lengthy narratives without hitting input limits.

  • Maximum 4,096 characters per request
  • Simple text splitting logic

Generate each segment

Send each chunk to the TTS API. Requests are processed sequentially, so longer texts may take time. Expect roughly real-time generation speed, with short pauses between requests.

Merge files for final output

Combine the generated audio files into one continuous file. This final audio file is ready for your audio guide hardware or mobile app.

  • Supports MP3, WAV, OPUS, FLAC

Data privacy and rights

Sound wave illustration for: Data privacy and rights

Your guide scripts remain private. We use your input text strictly to generate the audio file and do not store it for training or improving our models. Once the audio is created and returned to you, we do not keep a copy of the generated audio file either.

We retain only the request metadata—such as timestamp, character count, and status—to calculate your usage and generate accurate bills. This ensures transparency without compromising the content of your specific museum scripts or tourism materials.

Text not used for training

When you submit a script, the text is processed solely to produce the audio response. Unlike some services, we do not use your input text to train or improve our TTS engines. Your unique museum content remains yours alone.

Audio not stored by us

The generated audio file is returned directly in the API response. We do not store a permanent copy of your synthesized audio files on our servers. This means your specific guide recordings are not kept in our database after delivery.

Metadata kept for billing

We keep minimal metadata including the time of request, character count, and status code. This information is used exclusively to verify usage and calculate your charges accurately. It does not contain the actual text or audio content of your guides.

Licensing of voices

Sound wave illustration for: Licensing of voices

Our voice presets are grounded in real human recordings, ensuring natural and authentic speech patterns. Each voice is derived from licensed audio data, providing a high-quality foundation for your museum and tourism audio guides.

We value transparency regarding the origins of our synthetic voices. Understanding the source material helps you appreciate the authenticity and clarity of the audio generated for your visitors.

Based on LibriTTS-R data

The core of our voice models comes from the LibriTTS-R dataset. This extensive collection provides diverse vocal characteristics that we refine into our eight distinct presets.

  • High-quality source recordings
  • Diverse vocal profiles
  • Refined for clarity

CC BY 4.0 licensed source

Our source data is licensed under CC BY 4.0. This allows us to legally incorporate these recordings into our API while maintaining high standards for audio quality.

  • Open license terms
  • Legal clarity
  • High-quality standards

Public domain audiobook readings

Many of the base recordings originate from LibriVox public domain audiobooks. These readings offer rich, expressive narration styles that translate well into modern text-to-speech applications.

  • LibriVox origins
  • Expressive narration
  • Classic storytelling tone

Common limitations

Sound wave illustration for: Common limitations

Our API delivers complete audio files rather than streaming real-time audio. You receive the full MP3 or WAV after synthesis finishes. This approach is reliable for pre-generated museum guide content but does not support live, word-by-word playback during tours.

We do not support SSML markup for fine-grained control over pauses or emphasis. You also cannot clone your own voice or design a custom one. These features are not part of the current offering.

No streaming audio

Audio is generated as a complete file. You cannot stream partial audio chunks in real time. This is fine for standard audio guides but not for live interactive experiences.

Requests are processed sequentially, so long scripts may take time to synthesize fully before the file is available for download.

  • Best for pre-recorded guide content
  • Not suitable for live, word-by-word playback

No SSML support

The API accepts plain text input without SSML tags. You cannot specify precise pauses, pitch changes, or emphasis using markup. Control is limited to the speed parameter for overall tempo adjustments.

No custom voice cloning

We do not offer voice cloning or custom voice design. You must choose from the eight available presets. There is no option to upload a sample recording to create a unique voice model.

Getting started steps

Sound wave illustration for: Getting started steps

Start by visiting our /get-api-key/ page to create your account using either Google or an email address. This single step grants you access to your personal dashboard where your credentials and usage details are managed securely.

Sign up for an account

Register on the site using Google or email plus password. Accounts are created per site, giving you immediate access to your unique management dashboard.

Get your API key

Once registered, your dashboard displays your active API key and current balance. You can issue a new key at any time, which instantly invalidates the old one. Use this key in the Authorization header to authenticate your requests.

  • Key format: tts-...
  • One active key per account
  • Instant key rotation available

Generate your first audio

Send a POST request to the speech endpoint with your script and voice selection. The API returns the audio file directly, ready to be integrated into your museum guide app or website. Start with the free trial credit to test the workflow.

Listen to the 8 voices

Each sample below was generated by this API from the same sentence. Support your international visitors with clear, natural audio in their native tongue.

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. Node.js documentation
  3. Python documentation
  4. LibriTTS-R paper (arXiv 2305.18802)
  5. IETF RFC 7845: Ogg Opus
  6. LibriVox public-domain audiobooks
  7. curl documentation
  8. openai-node on GitHub
  9. TRON developer hub: TRC-20
  10. Creative Commons Attribution 4.0
  11. MDN: Web Speech API
  12. openai-python on GitHub

More on this site