rhvoice-ec

[rhvoice-ec]

Mateo was released

After our publication about SAPM (Sentence-Aware prosody modelling) we have been working better under it, expanding it to more contexts, and ensuring accuracy and stability.

This is the first Spanish TTS around the world that is capable to interpret the proper prosody of multi-context words, and it is very useful especially for screen reader users, hence they love our voice after two weeks of publication.

Regarding Screen reader or Visually Impaired users, we have news: we released Mateo, a voice with the slightly Ecuadorian accent, of the "sierra" region. This is just the start of this project.

The voice is available in open-source applications like Windows, Linux and the open-source runtime app for Android. Visit the .org site for more details.

However, we have a special release for the Play Store app, which is free only for Visual Impaired users. If you are not a Visual Impaired User running a screen reader like Talkback or Jieshuo, please help us with a subscription. This is useful to maintain the development of the main RHVoice project, in which this project is now connected, along with their servers and components. Any support is very appreciated!

[rhvoice-ec]

Our Spanish implementations using sentence-based accentuation according to context, something that no TTS systems do

We propose Sentence-Aware Prosody Modelling (SAPM), which consists in the control of the accentuation/stress in words that contains more than one context.

It’s important to remember that, in Spanish, there are unstressed words, especially in possessive adjectives, demonstratives, prepositions, articles, etc. For example, we can’t say “lós juegos” instead of “los juegos”. Fortunately, almost all Spanish TTS engines have these unstressed words into account in a basic way. However, there are some words that can contain more than one context. This is the case of “bajo”, which has four contexts: the bass instrument (tocar el bájo), the low/lower (muy bájo, tan bájo), lowering action (yo bájo, ya bájo) and “under” attribution (bajo la luz del sol). As you can notice, in the first three cases, “bajo” as stress in the “a”; however, in the last one, it is unstressed, which is the behavior by default. Other case is “sobre” with two contexts: paper object (el sóbre de papel) and “about” preposition (hablar sobre ti).

Please note that, we are forcing stress with acute vowels on purpose, to illustrate the goal of SAPM better. This is the way a normal speaker would interpret the text. Some other examples are: una tarde / cada úna, este chico / éste es mi amigo, como estos / yo cómo arroz.

We implemented SAPM in RHVoice successfully under a sentence-based backend, and according to Cedillo (2025), “this enriches the form of interpretation and communication of the sentence in TTS while reading texts, something that no TTS system has and, as a consequence, leads the user to confusion”. While for Cedillo likes this feature, we also want to hear people’s opinion to ensure that SAPM is a good approach. For that, we did a form recently, in which you can evaluate audio clips and select the one what you consider good.

[rhvoice-ec]

Merry christmas from the Ecuadorian voice

Another scenario demonstrating our voice. M. Cedillo, the first Ecuadorian Text-To-Speech voice, wrote his own reflection about christmas, and decided to share the message. In the last months, things had been complicated in Ecuador; however, we hope these are treated soon and come back to the "regular life".

Otro escenario que demuestra nuestra voz. M. Cedillo, la primera voz ecuatoriana de Texto a Voz, escribió su propia reflexión sobre la Navidad y decidió compartir el mensaje. En los últimos meses, las cosas se habían complicado en Ecuador; sin embargo, esperamos que pronto sean tratadas y volver a la "vida normal".

[rhvoice-ec]

Updates for the Ecuadorian Spanish voice

We are currently about to sell this voice and present it at companies or institutions in the future, but not without first polishing some previous phonetic and voice questions. Listen the audio so you know how we did it. We are ready to launch this project at this point.

[rhvoice-ec]

Ecuadorian Spanish demo

We built a new Ecuadorian Spanish voice, that has an accent of the Sierra region. This is a demonstration about how the voice sounds.

The recordings of the voice were made by a native speaker of the Sierra region, Matthew C., in the speaker's home studio.

We are planning to sell this voice in cash to start our entrepreneurship, and take this voice as a presentation of this project to institutions or companies in Ecuador, in the future.

[rhvoice-ec]

Our mission

Text-To-Speech (TTS) is a tool that helps people to communicate. There is a wide range of possibilities, such as blind people using screen readers, people with speech disabilities, means of transport that speaks information about a place, GPS systems and more.

TTS systems have been having innovation over the years; many people and companies have been adapted it for their needs, as well as worked in support for their languages. Today we have a lot of Text-To-Speech systems that offers his advantages. However, in the Ecuadorian land, no Text-To-Speech voices are available, hence they need to use a Mexican, a Colombian or a Castillian Spanish voice for their needs, so they will not be able to feel very identified with what they are communicating, as well as Kichwa speakers doesn't know how to use Text-To-Speech, because no voices for Kichwa and other languages are available, so they will have to communicate using a language other than their native language.

The mission of the RHVoice-ec project is facilitate communication for Ecuadorian speakers with voices in which they can identify, including indigenous languages. This way, we are creating a world of inclusion for this side.

RHVoice is a free, open-source TTS synthesizer. It supports many languages such as Russian, English, Portuguese, Czech, Esperanto, Servian, Nepali and much more. His voices are based in high quality recordings of natural speech databases, but they are small, because it uses a statistical parametric synthesis method. However, these voices sound pleasant to the ears and are fastest to respond. RHVoice also supports multiple platforms like Windows (sapi5 and NVDA screen reader), Linux, and Android.

Given the mission of this new project, we hope the support of different cultures in Ecuador to grow it. This project is still in the initial roots. Stay tuned!