[
    {
        "role": "user",
        "content": "Is%20there%20a%20preferable%20syntheciser%20for%20jaws%20and%20nvda%20that%20lets%20you%20seamlessly%20switch%20languages%20when%20the%20content%20changes%20language%3F"
    },
    {
        "role": "assistant",
        "content": "Here\u2019s a comparison of what JAWS and NVDA offer when it comes to switching speech synthesizers or \u201cvoices\u201d based on content language changes\u2014so you can pick the one that best meets your needs.\n\n---\n\n## NVDA: Multilingual Support and Automatic Switching\n\nNVDA supports automatic language switching quite well, as long as certain conditions are met.\n\n**Key features:**\n\n- **Automatic Language Switching**  \n  NVDA can automatically switch voice languages when reading text that\u2019s properly marked with a `lang` attribute (on web pages, Word documents, etc.). This works if your current synthesizer supports the target language. ([nvaccess.org](https://www.nvaccess.org/post/in-process-10th-april-2026/?utm_source=openai))\n\n- **Automatic Dialect Switching**  \n  If supported by your synthesizer, NVDA can also switch dialects automatically (e.g., U.S. vs. U.K. English) when the text is marked to specify those dialects. ([nvda.jp](https://www.nvda.jp/nvda2021.1jp/en/userGuide.html?utm_source=openai))\n\n- **Synthesizers that Support Switching**  \n  1. *eSpeak NG* \u2013 built in, supports 80+ languages, responsive. ([nvaccess.org](https://www.nvaccess.org/post/in-process-10th-april-2026/?utm_source=openai))  \n  2. *Windows OneCore voices* \u2013 natural-sounding, built into Windows 10/11; you\u2019ll need to install the specific voice packs for each language. ([nvaccess.org](https://www.nvaccess.org/post/in-process-10th-april-2026/?utm_source=openai))  \n  3. *Nuance Vocalizer for NVDA* \u2013 offers automatic language/dialect switching, with voices you can assign per language. ([vocalizer-nvda.com](https://www.vocalizer-nvda.com/docs/en/userguide.html?utm_source=openai))  \n\n**What to check:**\n\nTo ensure smooth automatic switching:\n\n- Make sure **Automatic Language Switching** is enabled in NVDA\u2019s speech settings. ([nvaccess-nvda.mintlify.app](https://nvaccess-nvda.mintlify.app/user-guide/speech-and-braille?utm_source=openai))  \n- Have the needed **voices installed** for each language you want to read. ([nvaccess.org](https://www.nvaccess.org/post/in-process-10th-april-2026/?utm_source=openai))  \n- Your content must have proper **language tags** (like `lang=\u201cfr\u201d` or Word\u2019s document language settings). ([nvda.jp](https://www.nvda.jp/nvda2024.4.1jp/en/changes.html?utm_source=openai))  \n\n---\n\n## JAWS: What It Offers (and Its Limitations)\n\nJAWS has some support for switching languages automatically, but it's more limited compared to NVDA, especially when it comes to switching *between different synthesizers*.\n\n**What works:**\n\n- **Voice Profiles with Multiple Voices**  \n  When using a synthesizer like *Vocalizer Expressive* with multiple languages installed, JAWS can switch voice profiles based on the language indicated in the document\u2014if voice packages for those languages are installed. ([visiondept.it](https://www.visiondept.it/surfup/Languages.htm?utm_source=openai))\n\n- **Language Detection Option**  \n  There\u2019s a setting called \u201cDetect Language\u201d (or \u201cDetect Languages\u201d) that must be turned on to enable JAWS to automatically use the correct voice profile when it encounters text in a different language. ([sensorysolutions.co.za](https://sensorysolutions.co.za/help-centre/switching-off-language-detection-in-jaws/?utm_source=openai))  \n\n**Where it falls short:**\n\n- **Synthesizer switching is restricted**  \n  JAWS can only switch voices *within* a single synthesizer. It won't automatically swap between synthesizers (e.g., switching from Eloquence to Vocalizer when the language changes) just based on text language. ([reddit.com](https://www.reddit.com/r/Blind/comments/n90pfj?utm_source=openai))  \n\n- **Strong dependence on correct markup**  \n  Just like NVDA, JAWS also requires that content is properly marked with language tags, both in web pages and documents. ([mail-archive.com](https://www.mail-archive.com/jaws-users-list%40jaws-users.com/msg92510.html?utm_source=openai))  \n\n---\n\n## Bottom Line: Which is Preferable?\n\nIf your priority is a seamless switch between languages whenever content changes, here's the recommendation:\n\n- **Go with NVDA**  \n  Particularly with a synthesizer like *Vocalizer Expressive*, *Windows OneCore*, or *eSpeak NG* that supports many languages and has the automatic language (and possibly dialect) switching feature. Provided your content is marked with the correct language metadata, NVDA will automatically switch voices mid-document\u2014no need for manual profile changes. ([nvaccess.org](https://www.nvaccess.org/post/in-process-10th-april-2026/?utm_source=openai))\n\n- **JAWS can work**, but with limitations.  \n  If you're willing to stick to a synthesizer supporting multiple languages (like Vocalizer Expressive) and ensure proper language tags, JAWS **can** switch voices automatically for language changes. But it won\u2019t switch synthesizers, and some user setups may still require manual switching. ([mail-archive.com](https://www.mail-archive.com/jaws-users-list%40jaws-users.com/msg92510.html?utm_source=openai))\n\n---\n\nIf you tell me **which languages** you use most (for example, English + Spanish + Chinese), and whether you're okay installing third-party voices or add-ons, I can suggest specific voice/synthesizer combinations optimized for *your* multilingual workflow."
    },
    {
        "role": "user",
        "content": "English%20and%20spannish...%20I%20am%20testing%20an%20exam%2C%20jaws%20with%20elpquence%20it%20was%20good%2C%20NVDA%20it%20did%20not%20switch%20to%20spannish%20accent."
    },
    {
        "role": "assistant",
        "content": "If you're using English and Spanish, and you want your screen reader to *automatically* switch accents or voices when the content language changes\u2014say, in an exam where some questions are in Spanish and others in English\u2014here\u2019s how JAWS and NVDA compare, plus specific settings and voices you can use to get the best results.\n\n---\n\n## NVDA Setup for English + Spanish\n\nNVDA *can* auto-switch voices when the language is marked in the content\u2014and it works best with certain synthesizers and settings. Here\u2019s what to do to make it work well.\n\n### What to install\n\n- Make sure you have **voices** installed for both English and Spanish. \n  - Use *Windows OneCore voices* (install Spanish voice pack via Windows Settings \u2192 Speech \u2192 Add voices). ([nvaccess.org](https://www.nvaccess.org/post/in-process-14th-january-2021/?utm_source=openai))  \n  - You can also use **Vocalizer Expressive** add-on, which supports automatic voice and accent switching. ([vocalizer-nvda.com](https://www.vocalizer-nvda.com/docs/en/userguide.html?utm_source=openai))\n\n- Check that the content has proper **language tags**:  \n  - On web pages, `<html lang=\"es\">` or `<span lang=\"es\">` helps  \n  - In Word or PDFs, check document or paragraph language settings. NVDA only detects correct language if the content is marked. ([vocalizer-nvda.com](https://www.vocalizer-nvda.com/docs/en/userguide.html?utm_source=openai))\n\n### Key NVDA settings\n\n- Go to **NVDA menu \u2192 Preferences \u2192 Speech Settings**.  \n  - Enable \u201cAutomatic Language Switching.\u201d NVDA will then attempt to switch voices when it sees text marked with a different language. ([nvaccess-nvda.mintlify.app](https://nvaccess-nvda.mintlify.app/configuration/speech-settings?utm_source=openai))  \n  - If you're using Vocalizer Expressive, there are extra options under its settings to map specific languages to specific voices. ([vocalizer-nvda.com](https://www.vocalizer-nvda.com/docs/en/userguide.html?utm_source=openai))\n\n- Also helpful:\n  - Enable \u201cDetect text language based on unicode characters\u201d (experimental) under Vocalizer to help in situations where language tags are missing. ([vocalizer-nvda.com](https://vocalizer-nvda.com/docs/en/userguide.pdf?utm_source=openai))\n\n### What might go wrong (and how to fix it)\n\n- If your synthesizer doesn\u2019t support Spanish (or Spanish voice isn\u2019t installed), NVDA will keep using the English voice even if Spanish text appears. ([reddit.com](https://www.reddit.com/r/accessibility/comments/pq20e9?utm_source=openai))  \n- If content isn\u2019t tagged with `lang=\"es\"` or equivalent, NVDA won\u2019t know it needs to switch. When that happens, \u201cdetect by unicode\u201d might help, but it\u2019s not 100% reliable.  \n- Some synthesizers may switch accent but still use English pronunciation rules\u2014so even though the \u201cvoice\u201d changes, some Spanish reads might sound \u201coff.\u201d\n\n---\n\n## JAWS with Eloquence or Vocalizer Expressive\n\nJAWS does support Spanish through **Eloquence** and **Vocalizer Expressive**, including Castilian and Latin American Spanish voices. ([support.freedomscientific.com](https://support.freedomscientific.com/support/technicalsupport/bulletin/1521?utm_source=openai))\n\n### What to do\n\n- Install the Spanish voices (Castilian, Latin American) via **JAWS** \u2192 *Options \u2192 Voices \u2192 Add/Remove Voices*. ([support.freedomscientific.com](https://support.freedomscientific.com/support/technicalsupport/bulletin/1521?utm_source=openai))  \n- Use the **Voice Profiles** feature (Insert + Ctrl + S) to set up separate profiles (one for English, one for Spanish) if you want manual control. ([support.freedomscientific.com](https://support.freedomscientific.com/support/technicalsupport/bulletin/1521?utm_source=openai))  \n- Enable \u201cDetect Languages/Detect Language\u201d features so JAWS can switch voice profiles automatically when it encounters a different language. Progress depends on how well the content is tagged.  \n\n### Limitations\n\n- JAWS generally can only switch voices *within* the same synthesizer, not between different synthesizers. So if Eloquence has both English + Spanish voices installed, it can switch between those when languages change; but it will not switch to a completely different speech engine automatically.  \n- If language tags are missing or content isn\u2019t well marked, automatic switching may not happen\u2014manual profile switching might be needed.\n\n---\n\n## Which Setup is Best for Your Exam\n\nBased on your test (English + Spanish), here\u2019s a suggestion:\n\n| Setup | Pros | Cons |\n|---|---|---|\n| **NVDA + Vocalizer Expressive** | Very good automatic switching, voices designed for different languages, highly customizable with mappings. | Need to install voices & ensure content is well tagged; some Spanish items still may pronounce some English-style words poorly. |\n| **JAWS + Eloquence (with Spanish voices installed)** | Strong Spanish voice quality, reliable for exam-like texts if Spanish voices installed, good for manual switching. | If content isn\u2019t tagged, switching might not happen; manual switching sometimes needed; accent consistency might depend on voice quality. |\n\n---\n\n## Recommendation\n\n- If you mostly want **automatic switching** and don't want to manually change profiles during the exam, **NVDA + Vocalizer Expressive** is likely the better choice\u2014if you get the Spanish voices installed and the content is correctly tagged.  \n- If you use **JAWS** and want good voice quality, make sure **both** Spanish voices are installed under Eloquence or Vocalizer, and that JAWS is set to detect language changes. But for absolute seamless switching, NVDA has a slight edge nowadays.\n\n---\n\nIf you want, I can step you through *exactly* how to set up NVDA or JAWS for your specific exam software (Word, PDF, a website, etc.), so that you can test and see the switching work before your exam. Do you want me to help with that configuration?"
    }
]