diff --git a/apps/desktop/src/app/settings/constants.ts b/apps/desktop/src/app/settings/constants.ts
index 2740b2d3ce8f..00cc65c7f711 100644
--- a/apps/desktop/src/app/settings/constants.ts
+++ b/apps/desktop/src/app/settings/constants.ts
@@ -387,7 +387,6 @@ export const FIELD_LABELS: Record<string, string> = defineFieldCopy({
       language: 'xAI Language',
       speed: 'xAI Playback Speed',
       autoSpeechTags: 'xAI Auto Speech Tags',
-      textNormalization: 'xAI Text Normalization',
       optimizeStreamingLatency: 'xAI Streaming Latency Optimization',
       sampleRate: 'xAI Sample Rate',
       bitRate: 'xAI Bit Rate'
@@ -500,7 +499,6 @@ export const FIELD_DESCRIPTIONS: Record<string, string> = defineFieldCopy({
       language: 'Spoken language code (e.g. en, pt-BR) or "auto" for auto-detection.',
       speed: 'Playback speed. 0.7 = slower, 1.0 = normal, 1.5 = faster.',
       autoSpeechTags: 'Let an LLM insert expressive audio tags ([laughing], [sighs]) into the script before synthesis.',
-      textNormalization: 'Convert numbers, abbreviations, and symbols to spoken form before synthesis.',
       optimizeStreamingLatency: 'Latency vs. quality trade-off. 0 = best quality, 2 = lowest latency.',
       sampleRate: 'Audio sample rate in Hz. Higher = better quality, larger files.',
       bitRate: 'MP3 bitrate in bps. Only applies when codec is mp3.'
@@ -606,7 +604,6 @@ export const SECTIONS: DesktopConfigSection[] = [
       'tts.xai.language',
       'tts.xai.speed',
       'tts.xai.auto_speech_tags',
-      'tts.xai.text_normalization',
       'tts.xai.optimize_streaming_latency',
       'tts.xai.sample_rate',
       'tts.xai.bit_rate',
diff --git a/apps/desktop/src/i18n/ja.ts b/apps/desktop/src/i18n/ja.ts
index 801936738328..d24e33ecff04 100644
--- a/apps/desktop/src/i18n/ja.ts
+++ b/apps/desktop/src/i18n/ja.ts
@@ -468,7 +468,6 @@ export const ja = defineLocale({
           language: 'xAI 言語',
           speed: '再生速度',
           autoSpeechTags: '自動音声タグ',
-          textNormalization: 'テキスト正規化',
           optimizeStreamingLatency: 'ストリーミング遅延最適化',
           sampleRate: 'サンプルレート',
           bitRate: 'ビットレート'
diff --git a/apps/desktop/src/i18n/zh-hant.ts b/apps/desktop/src/i18n/zh-hant.ts
index 72d43a3b2c24..4439465ad84a 100644
--- a/apps/desktop/src/i18n/zh-hant.ts
+++ b/apps/desktop/src/i18n/zh-hant.ts
@@ -457,7 +457,6 @@ export const zhHant = defineLocale({
           language: 'xAI 語言',
           speed: '播放速度',
           autoSpeechTags: '自動語音標籤',
-          textNormalization: '文字正規化',
           optimizeStreamingLatency: '串流延遲最佳化',
           sampleRate: '取樣率',
           bitRate: '位元率'
diff --git a/apps/desktop/src/i18n/zh.ts b/apps/desktop/src/i18n/zh.ts
index ec8b2ec58b83..c682cdae57a7 100644
--- a/apps/desktop/src/i18n/zh.ts
+++ b/apps/desktop/src/i18n/zh.ts
@@ -568,7 +568,6 @@ export const zh: Translations = {
           language: 'xAI 语言',
           speed: '播放速度',
           autoSpeechTags: '自动语音标签',
-          textNormalization: '文本规范化',
           optimizeStreamingLatency: '流式延迟优化',
           sampleRate: '采样率',
           bitRate: '比特率'
diff --git a/hermes_cli/config.py b/hermes_cli/config.py
index 4a01be9275cb..3cd431260c0e 100644
--- a/hermes_cli/config.py
+++ b/hermes_cli/config.py
@@ -2187,7 +2187,6 @@ DEFAULT_CONFIG = {
             "language": "en",  # BCP-47 code ("en", "pt-BR") or "auto"
             "speed": 1.0,  # 0.7–1.5, playback speed
             "auto_speech_tags": False,  # insert expressive audio tags via LLM rewrite
-            "text_normalization": False,  # normalize numbers/abbreviations/symbols to spoken form
             "optimize_streaming_latency": 0,  # 0–2, trades quality for lower latency
             "sample_rate": 24000,  # 22050 / 24000 / 44100 / 48000
             "bit_rate": 128000,  # MP3 bitrate; only applies when codec=mp3