Speech to text
Make every word usable.
Dictate into an editable transcript for notes and search. Select English, Amharic, or Afaan Oromo where your browser supports it.
Explore the toolsFrom Ethiopia. For the way we speak.
Turn spoken words and document images into text, and your knowledge into useful answers.
Speech, Amharic OCR, language models, and AI assistants with an Ethiopian language focus. Built to become part of your product.
Explore the tools. Bring us your language challenge.
ሰላም፣ እንዴት ነህ?
Capture. Understand. Respond.
The product toolkit
Start with one capability. Connect speech, document recognition, generation, and retrieval as your product grows.
Speech to text
Dictate into an editable transcript for notes and search. Select English, Amharic, or Afaan Oromo where your browser supports it.
Explore the toolsText to speech
Explore spoken interfaces, narration, and read-aloud experiences with an Ethiopian language focus.
Available device voicesText tools
Turn long notes into concise extracts, and explore language workflows for your application.
Explore the toolsDocument retrieval
Search a document with a question or keywords and find relevant passages with references to the source.
Explore the toolsAmharic OCR
Extract printed Amharic and English text from document images for editing, search, and knowledge workflows.
Explore the toolsImage extraction supports printed Amharic and English. Speech languages and voices depend on your device. Discuss your language and document requirements with our team.
A place to explore
Read an image, listen to your words, shorten a document, or find the passage you need. Bring your own input or start with an example.
Try it nowExtract editable text from a printed Amharic or English image.
PNG, JPEG or WebP. Up to 8 MB and 16 million pixels. Clear, printed text works best.
Your image is processed on your device. Language resources download on first use.
Starting dictation requests microphone access. Your browser's speech service may process audio remotely. Language support varies by browser.
Choose a voice available on your device. Some voices use an online speech service.
Extractive summary: selects sentences from your text and keeps their original wording.
Finds passages with matching words and cites their location in your document.
Choose an image or use an example to begin.
Your result will appear here.
Start with public or sample material. How your data is handled.
For people who build
Use the playground's text utilities in your own projects.
import {
summarizeText,
searchDocument
} from './text-tools.mjs';
const document = `
The library opens at 9 AM on weekdays.
Members can borrow books for two weeks.
The library closes at 5 PM.
`;
const summary = summarizeText(document, 2);
console.log(summary.text);
const passages = searchDocument(
document, 'When does the library open?'
);
console.log(passages);Save the downloaded module alongside your script. Runs in modern browsers and Node.js.
Measure what matters
Paste the expected text and your result to calculate word and character error rates for that pair.
Word error rate counts word substitutions, deletions, and insertions, divided by the number of reference words. Character error rate applies the same calculation to characters. Lower is better; insertions can push an error rate above 100%. Comparisons normalize Unicode, letter case, and whitespace. Word comparisons ignore punctuation; character comparisons include it. These results measure only the text pair you provide.
About EthioNeural
EthioNeural develops speech, document recognition, and language AI products with a focus on Ethiopian languages and the people who use them.
We bring open models, careful evaluation, and application engineering together. Our goal is to help teams turn language technology into something people can use every day.
Explore the public tools, then work with us on your language and document requirements. We build with upstream open models, browser capabilities, and application engineering.
Let's build something useful
Tell us what you want to build. We will discuss a demo, model fit, and the path to API access.
founder@ethioneural.comFor other questions:
contact@ethioneural.com