[
    {
        "role": "user",
        "content": "I%20think%20there%20is%20setting%20in%20nvda%20screen%20reader%20to%20turn%20off%20speaking%20of%20unicode.%20what%20is%20the%20setting%20and%20in%20what%20version%20of%20nvda%20was%20it%20introduced%3F"
    },
    {
        "role": "assistant",
        "content": "Here are the NVDA settings related to speaking Unicode characters or symbols, and the version in which they were introduced:\n\n---\n\n## \ud83d\udd0d Relevant Settings\n\nThere are a couple settings in NVDA that control whether or not NVDA will speak descriptions of Unicode symbols (including emoji), or apply Unicode normalization so that visually styled characters are read as their base character:\n\n1. **Include Unicode Consortium data (including emoji) when processing characters and symbols**  \n   - If this is ON, NVDA uses the Unicode Consortium\u2019s CLDR (Common Locale Data Repository) to speak symbol and emoji descriptions. If OFF, it avoids using those dictionary descriptions and may rely only on the speech synthesizer\u2019s built-in capabilities. ([nvda.jp](https://www.nvda.jp/nvda2021.1jp/en/userGuide.html?utm_source=openai))\n\n2. **Apply Unicode normalization to speech and braille output**  \n   - Normalization changes characters that are stylistic variants (like bold, italic, or other variant forms) into a standard or compatible form that NVDA can pronounce or render more consistently. This helps especially when a speech synthesizer doesn't have a pronunciation for some styled characters. ([nvaccess.org](https://www.nvaccess.org/post/nvda-2024-3/?utm_source=openai))\n\n---\n\n## \u2139 Version Introduced\n\n- The **\u201cApply Unicode normalization to speech and braille output\u201d** setting was introduced in NVDA **version 2024.3**. This feature was added so NVDA can handle characters that are unknown to certain synthesizers or braille tables by using compatible alternatives and to enable reading of equations in Word\u2019s equation editor. ([nvaccess.org](https://www.nvaccess.org/post/nvda-2024-3/?utm_source=openai))  \n- The setting **\u201cInclude Unicode Consortium data (including emoji)\u2026\u201c** already existed by 2021. It is accessible in the Speech Settings in several NVDA versions including 2021.3 and likely earlier. ([nvda.jp](https://www.nvda.jp/nvda2021.1jp/en/userGuide.html?utm_source=openai))\n\n---\n\n## \u2705 Summary\n\n- If you want to stop NVDA speaking Unicode-based descriptions (emoji, symbols, etc.), turn **off** **Include Unicode Consortium data (including emoji)\u2026** in the Speech \u2192 Characters & Symbols section.\n- If you also want NVDA to not convert variant characters via normalization (which might reduce how much \"extra\" Unicode detail gets spoken), ensure that **Apply Unicode normalization to speech and braille output** is configured as you prefer (ON or OFF). But this feature only exists from **NVDA 2024.3 onwards**. ([nvaccess.org](https://www.nvaccess.org/post/nvda-2024-3/?utm_source=openai))\n\n---\n\nIf you tell me the version of NVDA you\u2019re using, I can guide you specifically how to disable all Unicode speaking behavior in *your* version."
    }
]