Nothing saved yet. Documents you save manually will appear here, ready to download.
Use this guide when Ask AI is not answering. The four settings below are the important pieces: AI engine, API key, Endpoint, and Model.
Your API key is stored in this browser and sent directly to the provider. For a public website, a server-side proxy is safer because browser JavaScript cannot keep API keys secret.
Choose the service that will answer your questions.
If you do not have an API key yet, start with Browser AI if your browser supports it.
This is the secret credential supplied by your AI provider.
Never post your API key in screenshots, GitHub, or public HTML.
The endpoint is the web address where Voxoira sends the AI request.
If you get 404, CORS, or “failed to fetch”, the endpoint or provider access may be the problem.
The model is the specific AI you want the selected engine to run.
If your key works but you see “model not found”, check the model name first.
Quick visual references for the main Voxoira workflow and AI connection setup.
A browser-based workspace for voice dictation, translation, and AI-assisted text workflows.
Voxoira helps you turn speech into editable text, translate spoken or written content, and work with text using an AI assistant. The app is designed to keep the main workflow simple: speak, review, edit, translate, and ask AI.
Use browser speech recognition to create editable notes and documents. Voxoira includes language selection, microphone controls, formatting tools, autosave, copy, and download options.
Speak into the translation workspace, select languages, review the original and translated text, and use the available copy, download, and speech controls.
Ask AI can work with selected Voxoira text. Depending on your setup and browser support, it can use Browser AI or a third-party provider such as an OpenAI-compatible endpoint, Anthropic, or Gemini.
Voxoira is intended for everyday writing, notes, translation assistance, and text exploration. AI and speech results should always be reviewed before being used for important decisions.
For support, feedback, privacy requests, bug reports, or business inquiries, contact us by email.
Tell us what feature you were using, what browser/device you use, and what happened. Do not send passwords, API keys, or other secrets.
For questions about personal data, cookies, consent, or deletion requests, use the same support address and include “Privacy Request” in the subject.
This page explains how Voxoira may handle information when you use the website and its features.
Depending on the features you use, Voxoira may process text you type or dictate, translation input and output, AI prompts and responses, language preferences, and basic technical information required for the website to operate. Some settings may be stored locally in your browser.
When you activate dictation or translation, your browser may use its speech-recognition technology to process audio and return recognized text. Availability and processing behavior depend on your browser and its provider. Review your browser's privacy documentation for details.
If you choose an external AI provider, the text you submit may be sent to that provider to generate a response. The provider may process the request under its own terms and privacy policy. Voxoira does not control third-party processing.
Voxoira can store AI settings in your browser. API keys entered into the browser are sensitive credentials. Do not share them. For production deployments, a server-side integration can provide stronger protection than exposing provider credentials in client-side code.
Voxoira may use browser storage for preferences, consent choices, and application settings. Advertising technologies may use cookies or similar technologies when ads are enabled and consent or other applicable legal bases permit them.
If Google AdSense is enabled, Google and its partners may use cookies or similar identifiers to serve, measure, and personalize advertising as permitted by applicable settings and consent. Users should review the advertising and privacy information presented by the site's consent system.
You can control microphone permissions through your browser, change applicable cookie/advertising choices through the site's consent controls, and contact Voxoira about privacy questions or requests.
Reasonable technical measures should be used to protect information, but no online service can guarantee absolute security. Never submit passwords, payment details, API keys, or highly confidential information unless the feature specifically requires it and appropriate safeguards are in place.
Voxoira is not intended to encourage children to submit personal information. Parents or guardians should supervise use where appropriate.
This Privacy Policy may be updated as Voxoira, providers, advertising systems, or applicable requirements change. The current version published on this page is the version that applies.
For privacy questions or requests, contact contact@voxoira.com.
Answers to common questions about Voxoira's main features and AI connections.
Open Dictate, choose your language, allow microphone access when your browser asks, and start the microphone. Interim speech is shown live while you speak.
Check that the browser has microphone permission, the correct input device is selected, and another application is not blocking the microphone. HTTPS is recommended for production deployment.
Choose the source and target languages, start speech input, review the original text, and use the translation controls to review or speak the result.
External AI providers generally require authentication. Browser AI may work without a key when supported by the browser. Never publish a private provider key in client-side source code.
Usually the provider rejected the credential or account request. Check the API key, account permissions, billing/quota, and provider access.
It can indicate a network problem, incorrect endpoint, provider outage, or browser CORS restrictions. Verify the endpoint and provider documentation.
No. AI responses can be incorrect or incomplete. Review generated text before using it for legal, medical, financial, safety, or other high-impact decisions.
Email the site's privacy/support address listed on the Contact page. Do not include passwords or API keys in your message.
By accessing or using Voxoira, you agree to these Terms of Use. If you do not agree with these terms, please do not use or access the service.
By accessing this website or using Voxoira's dictation, translation, Ask AI, or related features, you agree to be bound by these Terms of Use, all applicable laws and regulations, and any applicable local laws. You are responsible for ensuring that your use of the service complies with the laws and rules that apply to you.
Voxoira is provided on an “as is” and “as available” basis. Voxoira makes no warranties, express or implied, and disclaims, to the extent permitted by law, implied warranties or conditions of merchantability, fitness for a particular purpose, non-infringement, accuracy, availability, or reliability.
Voxoira does not warrant that transcriptions, translations, AI responses, or other generated results will be accurate, complete, uninterrupted, or suitable for a particular purpose. Results can depend on your browser, microphone, speech-recognition service, translation service, AI provider, network connection, and other third-party technologies.
Voxoira is provided for reasonable and lawful use. You agree not to abuse, overload, disrupt, reverse engineer, interfere with, or attempt to circumvent technical or access controls of the service.
Users who abuse the service or violate these terms may have access restricted or permanently blocked where appropriate and permitted by law.
Ask AI may connect to third-party AI providers selected by you. Those providers may have their own terms, privacy policies, usage limits, pricing, and technical requirements. Voxoira is not responsible for third-party services or for their availability, policies, responses, or changes.
You are responsible for using API keys, endpoints, models, and provider accounts in accordance with the applicable provider's rules. Keep your API keys private and do not publish them in screenshots, source code, or public repositories.
You remain responsible for the text, recordings, prompts, documents, and other content that you enter, dictate, translate, or submit through Voxoira. You are also responsible for reviewing AI-generated or translated results before relying on them.
Do not use Voxoira as a substitute for professional legal, medical, financial, safety, or other specialized advice where professional judgment is required.
The Voxoira website, interface, branding, visual design, code, and other original materials are protected by applicable copyright, trademark, and other intellectual-property laws. You may not copy, reproduce, modify, redistribute, or commercially exploit protected materials except as permitted by law or with appropriate authorization.
To the maximum extent permitted by applicable law, Voxoira and its operators will not be liable for indirect, incidental, special, consequential, exemplary, or similar damages, including loss of data, loss of profits, loss of business, service interruption, or other losses arising from or related to your use of, or inability to use, the website or its materials and features.
Some jurisdictions do not allow certain warranty exclusions or limitations of liability, so some of these limitations may not apply to you.
Voxoira may update these Terms of Use from time to time as the service, features, or applicable requirements change. The updated version will apply when published on this website unless a different effective date is stated.
Important: These Terms of Use govern the use of Voxoira. By using voxoira.com, you agree to these terms. Voxoira recommends periodic legal review to ensure compliance with applicable laws and regulations.
Most people speak two to three times faster than they type. An online dictation tool lets you take advantage of that for first drafts, where speed of getting ideas down matters more than polish. Voxoira keeps the process simple: pick a language, press the microphone and speak.
Finished phrases are committed to the editor while the words still being recognized stay visible, so you always see what the tool is hearing. Text is auto-saved as you go, so an accidental refresh will not lose your work.
Emails and messages, blog drafts, meeting notes, journaling, study notes, and brainstorming all work well. Dictation is less suited to heavy formatting or code, where typing is more precise. A useful habit is to dictate the rough content first, then edit by keyboard for structure and wording.
Because everything runs in the browser, it works on any laptop or desktop with a modern browser and a microphone, with no software to install or keep updated.
Good dictation starts with the input. Choose the right microphone in your browser's site settings if you have more than one, and keep it a hand's width from your mouth. A laptop microphone works, but a headset reduces room echo and keyboard noise, which cuts errors noticeably.
Online dictation is not meant to replace careful proofreading. Treat the first pass as a draft, and read it aloud once before sending anything important.
No. Voxoira runs in the browser. You only need a microphone and permission to use it.
Yes. Select your language from the menu before starting. Recognition quality depends on the language and your browser.
Click in the editor and edit like any text box, or use Undo and Redo.
Voxoira sets no limit, but browsers can end a session after long silence. Press the microphone to continue.
In your browser's local storage on your device. Clearing site data removes it, so export anything important.
Dictation is faster than typing for most people, and it helps if typing is tiring or difficult. Speak in full sentences, say your punctuation clearly, and review the text before sending. There is nothing to install and no account to create.
Want to dictate straight into Gmail, Docs or a web form? Use the Voxoira Chrome extension. Working from recordings instead? See audio to text.
These options only affect the Dictate tab.