diff --git a/apps/desktop/src/app/settings/helpers.ts b/apps/desktop/src/app/settings/helpers.ts
index 548a481c39..a8b9b8df7f 100644
--- a/apps/desktop/src/app/settings/helpers.ts
+++ b/apps/desktop/src/app/settings/helpers.ts
@@ -177,7 +177,8 @@ export function voiceFieldVisible(key: string, config: HermesConfigRecord): bool
   const selected = String(getNested(config, `${domain}.provider`) ?? '')
   // Backend defaults when the key is unset: TTS → edge, STT → local.
   // An empty string used to hide every nested model field.
-  const fallback = domain === 'tts' ? 'edge' : domain === 'stt' ? 'local' : ''
+  const fallback = domain === 'tts' ? 'edge' : 'local'
+
   return provider === (selected || fallback)
 }
 
diff --git a/apps/desktop/src/app/settings/voice-provider-fields.test.ts b/apps/desktop/src/app/settings/voice-provider-fields.test.ts
index 27f5aafbae..0885d498bd 100644
--- a/apps/desktop/src/app/settings/voice-provider-fields.test.ts
+++ b/apps/desktop/src/app/settings/voice-provider-fields.test.ts
@@ -62,24 +62,9 @@ describe('voice field option coverage', () => {
     }
   })
 
-  it('suggests ElevenLabs v3 and current flash/turbo ids, not just multilingual v2', () => {
-    const models = ENUM_OPTIONS['tts.elevenlabs.model_id']
-
-    for (const model of [
-      'eleven_v3',
-      'eleven_ttv_v3',
-      'eleven_multilingual_v2',
-      'eleven_turbo_v2',
-      'eleven_turbo_v2_5',
-      'eleven_flash_v2',
-      'eleven_flash_v2_5'
-    ]) {
-      expect(models).toContain(model)
-    }
-  })
-
-  it('suggests gpt-transcribe as an OpenAI STT model', () => {
-    expect(ENUM_OPTIONS['stt.openai.model']).toContain('gpt-transcribe')
+  it('suggests the current ElevenLabs v3 model, not just the v2 trio', () => {
+    // Mirrors tools/tts_tool_delivery.py::ELEVENLABS_MODEL_MAX_TEXT_LENGTH.
+    expect(ENUM_OPTIONS['tts.elevenlabs.model_id']).toContain('eleven_v3')
   })
 
   it('keeps closed enums (devices, providers) out of the free-input set', () => {
diff --git a/contributors/emails/ramarivera@gmail.com b/contributors/emails/ramarivera@gmail.com
new file mode 100644
index 0000000000..92aade5771
--- /dev/null
+++ b/contributors/emails/ramarivera@gmail.com
@@ -0,0 +1 @@
+ramarivera
diff --git a/website/docs/user-guide/features/voice-mode.md b/website/docs/user-guide/features/voice-mode.md
index 4076c1e254..a33f6dde88 100644
--- a/website/docs/user-guide/features/voice-mode.md
+++ b/website/docs/user-guide/features/voice-mode.md
@@ -484,7 +484,7 @@ tts:
     voice: "en-US-AriaNeural"      # 322 voices, 74 languages
   elevenlabs:
     voice_id: "pNInz6obpgDQGcFmaJgB"    # Adam
-    model_id: "eleven_multilingual_v2"
+    model_id: "eleven_multilingual_v2"   # or eleven_v3, eleven_flash_v2_5, ... (Desktop Settings → Voice accepts any model id)
   openai:
     model: "gpt-4o-mini-tts"
     voice: "alloy"                 # alloy, echo, fable, onyx, nova, shimmer