diff --git a/Releases/v4.0.3/.claude/PAI-Install/engine/actions.ts b/Releases/v4.0.3/.claude/PAI-Install/engine/actions.ts
index 73ab5e212..5301324f7 100644
--- a/Releases/v4.0.3/.claude/PAI-Install/engine/actions.ts
+++ b/Releases/v4.0.3/.claude/PAI-Install/engine/actions.ts
@@ -10,7 +10,7 @@ import { homedir } from "os";
 import { join, basename } from "path";
 import type { InstallState, EngineEventHandler, DetectionResult } from "./types";
 import { PAI_VERSION, ALGORITHM_VERSION } from "./types";
-import { detectSystem, validateElevenLabsKey } from "./detect";
+import { detectSystem, validateElevenLabsKey, validateGoogleCloudKey } from "./detect";
 import { generateSettingsJson } from "./config-gen";
 
 /**
@@ -866,30 +866,87 @@ export async function runVoiceSetup(
 ): Promise<void> {
   await emit({ event: "step_start", step: "voice" });
 
-  // ── Collect ElevenLabs key if not already found ──
-  if (!state.collected.elevenLabsKey) {
-    await emit({ event: "progress", step: "voice", percent: 5, detail: "Searching for existing ElevenLabs key..." });
-    let elevenLabsKey = findExistingEnvKey("ELEVENLABS_API_KEY");
+  const paiDir = state.detection?.paiDir || join(homedir(), ".claude");
+  const settingsPath = join(paiDir, "settings.json");
+
+  // ── Choose TTS provider ──
+  await emit({ event: "progress", step: "voice", percent: 5, detail: "Choose your TTS provider..." });
+
+  const providerChoice = await getChoice("tts-provider", "Voice requires a TTS provider. Choose one:", [
+    { label: "ElevenLabs", value: "elevenlabs", description: "High-quality AI voices (10K free chars/month, paid plans available)" },
+    { label: "Google Cloud TTS", value: "google-cloud", description: "Neural2/WaveNet voices (4M free chars/month Standard, 1M WaveNet)" },
+    { label: "Skip voice for now", value: "skip", description: "You can configure voice later in settings.json" },
+  ]);
+
+  if (providerChoice === "skip") {
+    state.collected.ttsProvider = undefined;
+    await emit({ event: "message", content: "Voice skipped. You can add a TTS provider later in ~/.claude/settings.json" });
+    await emit({ event: "step_complete", step: "voice" });
+    return;
+  }
+
+  state.collected.ttsProvider = providerChoice as "elevenlabs" | "google-cloud";
 
-    if (elevenLabsKey) {
-      await emit({ event: "message", content: "Found existing ElevenLabs API key. Validating..." });
-      const result = await validateElevenLabsKey(elevenLabsKey);
+  // ── Collect API key for chosen provider ──
+  let hasApiKey = false;
+
+  if (providerChoice === "google-cloud") {
+    // ── Google Cloud TTS key collection ──
+    await emit({ event: "progress", step: "voice", percent: 10, detail: "Searching for existing Google Cloud key..." });
+    let googleKey = findExistingEnvKey("GOOGLE_CLOUD_API_KEY") || findExistingEnvKey("GOOGLE_API_KEY");
+
+    if (googleKey) {
+      await emit({ event: "message", content: "Found existing Google Cloud API key. Validating..." });
+      const result = await validateGoogleCloudKey(googleKey);
       if (result.valid) {
-        state.collected.elevenLabsKey = elevenLabsKey;
-        await emit({ event: "message", content: "Existing ElevenLabs API key is valid." });
+        state.collected.googleCloudKey = googleKey;
+        await emit({ event: "message", content: "Existing Google Cloud API key is valid." });
       } else {
         await emit({ event: "message", content: `Existing key invalid: ${result.error}.` });
-        elevenLabsKey = "";
+        googleKey = "";
       }
     }
 
-    if (!elevenLabsKey) {
-      const wantsVoice = await getChoice("voice-enable", "Voice requires an ElevenLabs API key. Get one free at elevenlabs.io", [
-        { label: "I have a key", value: "yes" },
-        { label: "Skip voice for now", value: "skip" },
-      ]);
+    if (!googleKey) {
+      const key = await getInput(
+        "google-cloud-key",
+        "Enter your Google Cloud API key:\nEnable the Text-to-Speech API at console.cloud.google.com",
+        "key",
+        "AIza..."
+      );
 
-      if (wantsVoice === "yes") {
+      if (key.trim()) {
+        await emit({ event: "progress", step: "voice", percent: 15, detail: "Validating Google Cloud key..." });
+        const result = await validateGoogleCloudKey(key.trim());
+        if (result.valid) {
+          state.collected.googleCloudKey = key.trim();
+          await emit({ event: "message", content: "Google Cloud API key verified." });
+        } else {
+          await emit({ event: "message", content: `Key validation failed: ${result.error}. Voice may not work until a valid key is provided.` });
+        }
+      }
+    }
+
+    hasApiKey = !!state.collected.googleCloudKey;
+  } else {
+    // ── ElevenLabs key collection (existing flow) ──
+    if (!state.collected.elevenLabsKey) {
+      await emit({ event: "progress", step: "voice", percent: 10, detail: "Searching for existing ElevenLabs key..." });
+      let elevenLabsKey = findExistingEnvKey("ELEVENLABS_API_KEY");
+
+      if (elevenLabsKey) {
+        await emit({ event: "message", content: "Found existing ElevenLabs API key. Validating..." });
+        const result = await validateElevenLabsKey(elevenLabsKey);
+        if (result.valid) {
+          state.collected.elevenLabsKey = elevenLabsKey;
+          await emit({ event: "message", content: "Existing ElevenLabs API key is valid." });
+        } else {
+          await emit({ event: "message", content: `Existing key invalid: ${result.error}.` });
+          elevenLabsKey = "";
+        }
+      }
+
+      if (!elevenLabsKey) {
         const key = await getInput(
           "elevenlabs-key",
           "Enter your ElevenLabs API key:",
@@ -904,109 +961,122 @@ export async function runVoiceSetup(
             state.collected.elevenLabsKey = key.trim();
             await emit({ event: "message", content: "ElevenLabs API key verified." });
           } else {
-            await emit({ event: "message", content: `Key validation failed: ${result.error}. Skipping voice setup.` });
+            await emit({ event: "message", content: `Key validation failed: ${result.error}. Voice may not work until a valid key is provided.` });
           }
         }
       }
     }
+
+    hasApiKey = !!state.collected.elevenLabsKey;
   }
 
-  const hasElevenLabsKey = !!state.collected.elevenLabsKey;
-  if (!hasElevenLabsKey) {
-    await emit({ event: "message", content: "No ElevenLabs key — voice server will use macOS text-to-speech as fallback. You can add a key later in ~/.config/PAI/.env" });
+  if (!hasApiKey) {
+    await emit({ event: "message", content: `No ${providerChoice === "google-cloud" ? "Google Cloud" : "ElevenLabs"} key — voice will need manual configuration. Add the key to ~/.config/PAI/.env` });
   }
 
-  // ── Start voice server (works with or without ElevenLabs key) ──
-  const paiDir = state.detection?.paiDir || join(homedir(), ".claude");
+  // ── Start voice server ──
   await emit({ event: "progress", step: "voice", percent: 25, detail: "Starting voice server..." });
   const voiceServerReady = await startVoiceServer(paiDir, emit);
 
-  // ── Digital Assistant Voice selection ──
-  await emit({ event: "progress", step: "voice", percent: 40, detail: "Checking for existing voice configuration..." });
+  // ── Voice selection (ElevenLabs only — Google Cloud uses Neural2-D default) ──
+  let selectedVoiceId = "";
 
-  const voiceIds: Record<string, string> = {
-    male: "pNInz6obpgDQGcFmaJgB",
-    female: "21m00Tcm4TlvDq8ikWAM",
-  };
+  if (providerChoice === "google-cloud") {
+    // Google Cloud: use default Neural2-D voice, configurable later in settings.json
+    selectedVoiceId = ""; // Not applicable for Google Cloud
+    await emit({ event: "message", content: "Google Cloud TTS will use Neural2-D voice (en-US). Configurable in settings.json → daidentity.googleCloudVoice" });
+  } else {
+    // ElevenLabs: existing voice selection flow
+    await emit({ event: "progress", step: "voice", percent: 40, detail: "Checking for existing voice configuration..." });
 
-  let selectedVoiceId: string;
+    const voiceIds: Record<string, string> = {
+      male: "pNInz6obpgDQGcFmaJgB",
+      female: "21m00Tcm4TlvDq8ikWAM",
+    };
 
-  // Check for existing voice config from previous installations
-  const existingVoice = findExistingVoiceConfig();
+    // Check for existing voice config from previous installations
+    const existingVoice = findExistingVoiceConfig();
 
-  if (existingVoice) {
-    const sourceLabel = existingVoice.aiName
-      ? `${existingVoice.aiName}'s voice (${existingVoice.voiceId.substring(0, 8)}...)`
-      : `Voice ID ${existingVoice.voiceId.substring(0, 8)}...`;
-    await emit({ event: "message", content: `Found existing voice configuration from ~/${existingVoice.source}` });
+    if (existingVoice) {
+      const sourceLabel = existingVoice.aiName
+        ? `${existingVoice.aiName}'s voice (${existingVoice.voiceId.substring(0, 8)}...)`
+        : `Voice ID ${existingVoice.voiceId.substring(0, 8)}...`;
+      await emit({ event: "message", content: `Found existing voice configuration from ~/${existingVoice.source}` });
 
-    const useExisting = await getChoice("voice-existing", `Your DA was using: ${sourceLabel}. Use the same voice?`, [
-      { label: "Yes, keep this voice", value: "keep", description: `Voice ID: ${existingVoice.voiceId}` },
-      { label: "No, pick a new voice", value: "new", description: "Choose from presets or enter a custom ID" },
-    ]);
+      const useExisting = await getChoice("voice-existing", `Your DA was using: ${sourceLabel}. Use the same voice?`, [
+        { label: "Yes, keep this voice", value: "keep", description: `Voice ID: ${existingVoice.voiceId}` },
+        { label: "No, pick a new voice", value: "new", description: "Choose from presets or enter a custom ID" },
+      ]);
 
-    if (useExisting === "keep") {
-      selectedVoiceId = existingVoice.voiceId;
-      state.collected.voiceType = "custom";
-      state.collected.customVoiceId = selectedVoiceId;
-    } else {
-      // Fall through to voice selection below
-      selectedVoiceId = "";
+      if (useExisting === "keep") {
+        selectedVoiceId = existingVoice.voiceId;
+        state.collected.voiceType = "custom";
+        state.collected.customVoiceId = selectedVoiceId;
+      }
     }
-  } else {
-    selectedVoiceId = "";
-  }
 
-  // Voice selection (if not using existing)
-  if (!selectedVoiceId) {
-    await emit({ event: "progress", step: "voice", percent: 45, detail: "Choose your Digital Assistant's voice..." });
-
-    const voiceType = await getChoice("voice-type", "Digital Assistant Voice — Choose a voice for your AI assistant:", [
-      { label: "Female (Rachel)", value: "female", description: "Warm, articulate female voice" },
-      { label: "Male (Adam)", value: "male", description: "Clear, confident male voice" },
-      { label: "Custom Voice ID", value: "custom", description: "Enter your own ElevenLabs voice ID" },
-    ]);
-
-    if (voiceType === "custom") {
-      const customId = await getInput(
-        "custom-voice-id",
-        "Enter your ElevenLabs Voice ID:\nFind it at: elevenlabs.io/app/voice-library → Your voice → Voice ID",
-        "text",
-        "e.g., s3TPKV1kjDlVtZbl4Ksh"
-      );
-      selectedVoiceId = customId.trim() || voiceIds.female;
-      state.collected.voiceType = "custom";
-      state.collected.customVoiceId = selectedVoiceId;
-    } else {
-      selectedVoiceId = voiceIds[voiceType] || voiceIds.female;
-      state.collected.voiceType = voiceType as any;
+    // Voice selection (if not using existing)
+    if (!selectedVoiceId) {
+      await emit({ event: "progress", step: "voice", percent: 45, detail: "Choose your Digital Assistant's voice..." });
+
+      const voiceType = await getChoice("voice-type", "Digital Assistant Voice — Choose a voice for your AI assistant:", [
+        { label: "Female (Rachel)", value: "female", description: "Warm, articulate female voice" },
+        { label: "Male (Adam)", value: "male", description: "Clear, confident male voice" },
+        { label: "Custom Voice ID", value: "custom", description: "Enter your own ElevenLabs voice ID" },
+      ]);
+
+      if (voiceType === "custom") {
+        const customId = await getInput(
+          "custom-voice-id",
+          "Enter your ElevenLabs Voice ID:\nFind it at: elevenlabs.io/app/voice-library → Your voice → Voice ID",
+          "text",
+          "e.g., s3TPKV1kjDlVtZbl4Ksh"
+        );
+        selectedVoiceId = customId.trim() || voiceIds.female;
+        state.collected.voiceType = "custom";
+        state.collected.customVoiceId = selectedVoiceId;
+      } else {
+        selectedVoiceId = voiceIds[voiceType] || voiceIds.female;
+        state.collected.voiceType = voiceType as any;
+      }
     }
   }
 
-  // ── Update settings.json with voice ID ──
+  // ── Update settings.json with voice + provider config ──
   await emit({ event: "progress", step: "voice", percent: 60, detail: "Saving voice configuration..." });
-  const settingsPath = join(paiDir, "settings.json");
 
   if (existsSync(settingsPath)) {
     try {
       const settings = JSON.parse(readFileSync(settingsPath, "utf-8"));
       if (settings.daidentity) {
-        settings.daidentity.voiceId = selectedVoiceId;
-        settings.daidentity.voices = settings.daidentity.voices || {};
-        settings.daidentity.voices.main = {
-          voiceId: selectedVoiceId,
-          stability: 0.35,
-          similarityBoost: 0.80,
-          style: 0.90,
-          speed: 1.1,
-        };
-        settings.daidentity.voices.algorithm = {
-          voiceId: selectedVoiceId,
-          stability: 0.35,
-          similarityBoost: 0.80,
-          style: 0.90,
-          speed: 1.1,
-        };
+        settings.daidentity.ttsProvider = providerChoice;
+
+        if (providerChoice === "google-cloud") {
+          settings.daidentity.googleCloudVoice = {
+            languageCode: "en-US",
+            voiceName: "en-US-Neural2-D",
+            voiceType: "NEURAL2",
+            speakingRate: 1.0,
+            pitch: 0.0,
+          };
+        } else {
+          settings.daidentity.voiceId = selectedVoiceId;
+          settings.daidentity.voices = settings.daidentity.voices || {};
+          settings.daidentity.voices.main = {
+            voiceId: selectedVoiceId,
+            stability: 0.35,
+            similarityBoost: 0.80,
+            style: 0.90,
+            speed: 1.1,
+          };
+          settings.daidentity.voices.algorithm = {
+            voiceId: selectedVoiceId,
+            stability: 0.35,
+            similarityBoost: 0.80,
+            style: 0.90,
+            speed: 1.1,
+          };
+        }
       }
       writeFileSync(settingsPath, JSON.stringify(settings, null, 2));
       await emit({ event: "message", content: "Voice settings saved to settings.json." });
@@ -1015,17 +1085,26 @@ export async function runVoiceSetup(
     }
   }
 
-  // ── Save ElevenLabs key to .env (if provided) ──
-  if (hasElevenLabsKey) {
+  // ── Save API key to .env ──
+  if (hasApiKey) {
     const configDir = state.detection?.configDir || join(homedir(), ".config", "PAI");
     const envPath = join(configDir, ".env");
     if (!existsSync(configDir)) mkdirSync(configDir, { recursive: true });
 
     let envContent = existsSync(envPath) ? readFileSync(envPath, "utf-8") : "";
-    if (envContent.includes("ELEVENLABS_API_KEY=")) {
-      envContent = envContent.replace(/ELEVENLABS_API_KEY=.*/, `ELEVENLABS_API_KEY=${state.collected.elevenLabsKey}`);
-    } else {
-      envContent = envContent.trim() + `\nELEVENLABS_API_KEY=${state.collected.elevenLabsKey}\n`;
+
+    if (providerChoice === "google-cloud" && state.collected.googleCloudKey) {
+      if (envContent.includes("GOOGLE_CLOUD_API_KEY=")) {
+        envContent = envContent.replace(/GOOGLE_CLOUD_API_KEY=.*/, `GOOGLE_CLOUD_API_KEY=${state.collected.googleCloudKey}`);
+      } else {
+        envContent = envContent.trim() + `\nGOOGLE_CLOUD_API_KEY=${state.collected.googleCloudKey}\n`;
+      }
+    } else if (state.collected.elevenLabsKey) {
+      if (envContent.includes("ELEVENLABS_API_KEY=")) {
+        envContent = envContent.replace(/ELEVENLABS_API_KEY=.*/, `ELEVENLABS_API_KEY=${state.collected.elevenLabsKey}`);
+      } else {
+        envContent = envContent.trim() + `\nELEVENLABS_API_KEY=${state.collected.elevenLabsKey}\n`;
+      }
     }
     writeFileSync(envPath, envContent.trim() + "\n", { mode: 0o600 });
 
@@ -1053,32 +1132,38 @@ export async function runVoiceSetup(
       try {
         const aiName = state.collected.aiName || "PAI";
         const userName = state.collected.principalName || "there";
+
+        // Build test payload based on provider
+        const testPayload: Record<string, any> = {
+          message: `Hello ${userName}, this is ${aiName}. My voice system is online and ready to assist you.`,
+        };
+        if (providerChoice === "elevenlabs" && selectedVoiceId) {
+          testPayload.voice_id = selectedVoiceId;
+          testPayload.voice_settings = { stability: 0.35, similarity_boost: 0.80, style: 0.90, speed: 1.1, use_speaker_boost: true };
+        }
+
         const testRes = await fetch("http://localhost:8888/notify", {
           method: "POST",
           headers: { "Content-Type": "application/json" },
-          body: JSON.stringify({
-            message: `Hello ${userName}, this is ${aiName}. My voice system is online and ready to assist you.`,
-            voice_id: selectedVoiceId,
-            voice_settings: { stability: 0.35, similarity_boost: 0.80, style: 0.90, speed: 1.1, use_speaker_boost: true },
-          }),
-          signal: AbortSignal.timeout(10000),
+          body: JSON.stringify(testPayload),
+          signal: AbortSignal.timeout(15000),
         });
         if (testRes.ok) {
           await emit({ event: "message", content: `Voice test sent — listen for ${aiName} speaking...`, speak: false });
 
           // Ask user to confirm they heard it and like it
-          const confirm = await getChoice("voice-confirm", "Did you hear the voice? Does it sound good?", [
+          const confirmChoices = [
             { label: "Sounds great!", value: "yes" },
-            { label: "Pick a different voice", value: "change" },
+            ...(providerChoice === "elevenlabs" ? [{ label: "Pick a different voice", value: "change" }] : []),
             { label: "Skip voice for now", value: "skip" },
-          ]);
+          ];
+          const confirm = await getChoice("voice-confirm", "Did you hear the voice? Does it sound good?", confirmChoices);
 
-          if (confirm === "yes") {
-            voiceConfirmed = true;
-          } else if (confirm === "skip") {
+          if (confirm === "yes" || confirm === "skip") {
             voiceConfirmed = true;
           } else {
-            // Let them pick again
+            // ElevenLabs: let them pick again
+            const voiceIds: Record<string, string> = { male: "pNInz6obpgDQGcFmaJgB", female: "21m00Tcm4TlvDq8ikWAM" };
             const newVoice = await getChoice("voice-type-retry", "Choose a different voice:", [
               { label: "Female (Rachel)", value: "female", description: "Warm, articulate female voice" },
               { label: "Male (Adam)", value: "male", description: "Clear, confident male voice" },
@@ -1112,9 +1197,11 @@ export async function runVoiceSetup(
     }
   }
 
-  const voiceLabel = state.collected.voiceType === "custom"
-    ? `Custom voice (${selectedVoiceId.substring(0, 8)}...)`
-    : state.collected.voiceType || "default";
-  await emit({ event: "message", content: `Digital Assistant voice configured: ${voiceLabel}` });
+  const voiceLabel = providerChoice === "google-cloud"
+    ? "Google Cloud TTS (Neural2-D)"
+    : state.collected.voiceType === "custom"
+      ? `Custom voice (${selectedVoiceId.substring(0, 8)}...)`
+      : state.collected.voiceType || "default";
+  await emit({ event: "message", content: `Digital Assistant voice configured: ${voiceLabel} (${providerChoice})` });
   await emit({ event: "step_complete", step: "voice" });
 }
diff --git a/Releases/v4.0.3/.claude/PAI-Install/engine/config-gen.ts b/Releases/v4.0.3/.claude/PAI-Install/engine/config-gen.ts
index 51789a6a3..cdccdeb68 100644
--- a/Releases/v4.0.3/.claude/PAI-Install/engine/config-gen.ts
+++ b/Releases/v4.0.3/.claude/PAI-Install/engine/config-gen.ts
@@ -35,6 +35,10 @@ export function generateSettingsJson(config: PAIConfig): Record<string, any> {
       fullName: `${config.aiName} — Personal AI`,
       displayName: config.aiName.toUpperCase(),
       color: "#3B82F6",
+      ...(config.ttsProvider ? { ttsProvider: config.ttsProvider } : {}),
+      ...(config.ttsProvider === "google-cloud" && config.googleCloudVoice
+        ? { googleCloudVoice: config.googleCloudVoice }
+        : {}),
       voices: {
         main: {
           voiceId,
diff --git a/Releases/v4.0.3/.claude/PAI-Install/engine/detect.ts b/Releases/v4.0.3/.claude/PAI-Install/engine/detect.ts
index 3dc25f3bf..a5cb29a10 100644
--- a/Releases/v4.0.3/.claude/PAI-Install/engine/detect.ts
+++ b/Releases/v4.0.3/.claude/PAI-Install/engine/detect.ts
@@ -100,7 +100,8 @@ function detectExisting(
     try {
       const envContent = readFileSync(envPath, "utf-8");
       result.elevenLabsKeyFound = envContent.includes("ELEVENLABS_API_KEY=");
-      result.hasApiKeys = result.elevenLabsKeyFound;
+      const googleKeyFound = envContent.includes("GOOGLE_CLOUD_API_KEY=") || envContent.includes("GOOGLE_API_KEY=");
+      result.hasApiKeys = result.elevenLabsKeyFound || googleKeyFound;
     } catch {
       // Permission denied or other error
     }
@@ -166,3 +167,21 @@ export async function validateElevenLabsKey(key: string): Promise<{ valid: boole
     return { valid: false, error: e.message || "Network error" };
   }
 }
+
+/**
+ * Validate a Google Cloud API key by listing available TTS voices.
+ */
+export async function validateGoogleCloudKey(key: string): Promise<{ valid: boolean; error?: string }> {
+  try {
+    const res = await fetch(
+      `https://texttospeech.googleapis.com/v1/voices?key=${key}`,
+      { signal: AbortSignal.timeout(10000) }
+    );
+
+    if (res.ok) return { valid: true };
+    const body = await res.text().catch(() => "");
+    return { valid: false, error: `HTTP ${res.status}${body ? `: ${body.slice(0, 100)}` : ""}` };
+  } catch (e: any) {
+    return { valid: false, error: e.message || "Network error" };
+  }
+}
diff --git a/Releases/v4.0.3/.claude/PAI-Install/engine/steps.ts b/Releases/v4.0.3/.claude/PAI-Install/engine/steps.ts
index 7d5e14d16..10c6bd4b6 100644
--- a/Releases/v4.0.3/.claude/PAI-Install/engine/steps.ts
+++ b/Releases/v4.0.3/.claude/PAI-Install/engine/steps.ts
@@ -25,7 +25,7 @@ export const STEPS: StepDefinition[] = [
   {
     id: "api-keys",
     name: "API Keys",
-    description: "Find or collect ElevenLabs API key for voice features",
+    description: "Find or collect API keys for voice features",
     number: 3,
     required: true,
     dependsOn: ["prerequisites"],
@@ -57,7 +57,7 @@ export const STEPS: StepDefinition[] = [
   {
     id: "voice",
     name: "Digital Assistant Voice",
-    description: "Configure ElevenLabs key, select voice, start voice server, and test",
+    description: "Choose TTS provider, configure API key, select voice, and test",
     number: 7,
     required: true,
     dependsOn: ["configuration"],
diff --git a/Releases/v4.0.3/.claude/PAI-Install/engine/types.ts b/Releases/v4.0.3/.claude/PAI-Install/engine/types.ts
index 0e93734d8..8ab0bb569 100644
--- a/Releases/v4.0.3/.claude/PAI-Install/engine/types.ts
+++ b/Releases/v4.0.3/.claude/PAI-Install/engine/types.ts
@@ -79,6 +79,8 @@ export interface InstallState {
   // Collected data
   collected: {
     elevenLabsKey?: string;
+    googleCloudKey?: string;
+    ttsProvider?: "elevenlabs" | "google-cloud";
     principalName?: string;
     timezone?: string;
     aiName?: string;
@@ -112,6 +114,14 @@ export interface PAIConfig {
   temperatureUnit?: "fahrenheit" | "celsius";
   voiceType?: string;
   voiceId?: string;
+  ttsProvider?: "elevenlabs" | "google-cloud";
+  googleCloudVoice?: {
+    languageCode: string;
+    voiceName: string;
+    voiceType: string;
+    speakingRate: number;
+    pitch: number;
+  };
   paiDir: string;
   configDir: string;
 }
@@ -127,7 +137,7 @@ export type ServerMessage =
   | { type: "input_request"; id: string; prompt: string; inputType: "text" | "password" | "key"; placeholder?: string }
   | { type: "choice_request"; id: string; prompt: string; choices: { label: string; value: string; description?: string }[] }
   | { type: "progress"; step: StepId; percent: number; detail: string }
-  | { type: "voice_enabled"; enabled: boolean; mode: "elevenlabs" | "browser" | "none" }
+  | { type: "voice_enabled"; enabled: boolean; mode: "elevenlabs" | "google-cloud" | "browser" | "none" }
   | { type: "install_complete"; success: boolean; summary: InstallSummary }
   | { type: "validation_result"; checks: ValidationCheck[] }
   | { type: "error"; message: string; step?: StepId };
diff --git a/Releases/v4.0.3/.claude/VoiceServer/install.sh b/Releases/v4.0.3/.claude/VoiceServer/install.sh
index 15b6c83e0..dff7b2fd4 100755
--- a/Releases/v4.0.3/.claude/VoiceServer/install.sh
+++ b/Releases/v4.0.3/.claude/VoiceServer/install.sh
@@ -1,7 +1,7 @@
 #!/bin/bash
 
 # Voice Server Installation Script
-# This script installs the voice server as a macOS service
+# Installs the voice server as a macOS LaunchAgent or Linux systemd user service
 
 set -e
 
@@ -14,13 +14,27 @@ NC='\033[0m' # No Color
 
 # Configuration
 SCRIPT_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )"
-SERVICE_NAME="com.pai.voice-server"
-PLIST_PATH="$HOME/Library/LaunchAgents/${SERVICE_NAME}.plist"
-LOG_PATH="$HOME/Library/Logs/pai-voice-server.log"
+OS_TYPE="$(uname -s)"
 ENV_FILE="$HOME/.env"
 
+# Platform-specific paths
+if [ "$OS_TYPE" = "Darwin" ]; then
+    SERVICE_NAME="com.pai.voice-server"
+    PLIST_PATH="$HOME/Library/LaunchAgents/${SERVICE_NAME}.plist"
+    LOG_PATH="$HOME/Library/Logs/pai-voice-server.log"
+elif [ "$OS_TYPE" = "Linux" ]; then
+    SERVICE_NAME="pai-voice-server"
+    SYSTEMD_DIR="$HOME/.config/systemd/user"
+    SERVICE_PATH="${SYSTEMD_DIR}/${SERVICE_NAME}.service"
+    LOG_PATH="$HOME/.config/pai/voice-server.log"
+else
+    echo -e "${RED}X Unsupported platform: $OS_TYPE${NC}"
+    echo "  PAI Voice Server supports macOS and Linux."
+    exit 1
+fi
+
 echo -e "${BLUE}=====================================================${NC}"
-echo -e "${BLUE}     PAI Voice Server Installation${NC}"
+echo -e "${BLUE}     PAI Voice Server Installation ($OS_TYPE)${NC}"
 echo -e "${BLUE}=====================================================${NC}"
 echo
 
@@ -34,52 +48,97 @@ if ! command -v bun &> /dev/null; then
 fi
 echo -e "${GREEN}OK Bun is installed${NC}"
 
-# Check for existing installation
-if launchctl list | grep -q "$SERVICE_NAME" 2>/dev/null; then
-    echo -e "${YELLOW}! Voice server is already installed${NC}"
-    read -p "Do you want to reinstall? (y/n): " -n 1 -r
-    echo
-    if [[ $REPLY =~ ^[Yy]$ ]]; then
-        echo -e "${YELLOW}> Stopping existing service...${NC}"
-        launchctl unload "$PLIST_PATH" 2>/dev/null || true
-        echo -e "${GREEN}OK Existing service stopped${NC}"
+# Linux: check for audio player
+if [ "$OS_TYPE" = "Linux" ]; then
+    if command -v mpv &> /dev/null; then
+        echo -e "${GREEN}OK Audio player: mpv${NC}"
+    elif command -v ffplay &> /dev/null; then
+        echo -e "${GREEN}OK Audio player: ffplay${NC}"
+    elif command -v aplay &> /dev/null; then
+        echo -e "${GREEN}OK Audio player: aplay${NC}"
     else
-        echo "Installation cancelled"
-        exit 0
+        echo -e "${YELLOW}! No audio player found (mpv or ffplay recommended)${NC}"
+        echo "  Install one: sudo apt install mpv  OR  sudo apt install ffmpeg"
+    fi
+    if command -v notify-send &> /dev/null; then
+        echo -e "${GREEN}OK Desktop notifications: notify-send${NC}"
+    else
+        echo -e "${YELLOW}! notify-send not found — desktop notifications will be skipped${NC}"
+        echo "  Install: sudo apt install libnotify-bin"
     fi
 fi
 
-# Check for ElevenLabs configuration
-echo -e "${YELLOW}> Checking ElevenLabs configuration...${NC}"
-if [ -f "$ENV_FILE" ] && grep -q "ELEVENLABS_API_KEY=" "$ENV_FILE"; then
-    API_KEY=$(grep "ELEVENLABS_API_KEY=" "$ENV_FILE" | cut -d'=' -f2)
-    if [ "$API_KEY" != "your_api_key_here" ] && [ -n "$API_KEY" ]; then
-        echo -e "${GREEN}OK ElevenLabs API key configured${NC}"
-        ELEVENLABS_CONFIGURED=true
-    else
-        echo -e "${YELLOW}! ElevenLabs API key not configured${NC}"
-        echo "  Voice server will use macOS 'say' command as fallback"
-        ELEVENLABS_CONFIGURED=false
+# Check for existing installation
+if [ "$OS_TYPE" = "Darwin" ]; then
+    if launchctl list | grep -q "$SERVICE_NAME" 2>/dev/null; then
+        echo -e "${YELLOW}! Voice server is already installed${NC}"
+        read -p "Do you want to reinstall? (y/n): " -n 1 -r
+        echo
+        if [[ $REPLY =~ ^[Yy]$ ]]; then
+            echo -e "${YELLOW}> Stopping existing service...${NC}"
+            launchctl unload "$PLIST_PATH" 2>/dev/null || true
+            echo -e "${GREEN}OK Existing service stopped${NC}"
+        else
+            echo "Installation cancelled"
+            exit 0
+        fi
+    fi
+elif [ "$OS_TYPE" = "Linux" ]; then
+    if systemctl --user is-active "$SERVICE_NAME" &>/dev/null; then
+        echo -e "${YELLOW}! Voice server is already installed${NC}"
+        read -p "Do you want to reinstall? (y/n): " -n 1 -r
+        echo
+        if [[ $REPLY =~ ^[Yy]$ ]]; then
+            echo -e "${YELLOW}> Stopping existing service...${NC}"
+            systemctl --user stop "$SERVICE_NAME" 2>/dev/null || true
+            systemctl --user disable "$SERVICE_NAME" 2>/dev/null || true
+            echo -e "${GREEN}OK Existing service stopped${NC}"
+        else
+            echo "Installation cancelled"
+            exit 0
+        fi
     fi
-else
-    echo -e "${YELLOW}! No ElevenLabs configuration found${NC}"
-    echo "  Voice server will use macOS 'say' command as fallback"
-    ELEVENLABS_CONFIGURED=false
 fi
 
-if [ "$ELEVENLABS_CONFIGURED" = false ]; then
+# Check for TTS API configuration
+echo -e "${YELLOW}> Checking TTS configuration...${NC}"
+TTS_CONFIGURED=false
+TTS_PROVIDER="none"
+
+if [ -f "$ENV_FILE" ]; then
+    if grep -q "ELEVENLABS_API_KEY=" "$ENV_FILE"; then
+        API_KEY=$(grep "ELEVENLABS_API_KEY=" "$ENV_FILE" | cut -d'=' -f2)
+        if [ "$API_KEY" != "your_api_key_here" ] && [ -n "$API_KEY" ]; then
+            echo -e "${GREEN}OK ElevenLabs API key configured${NC}"
+            TTS_CONFIGURED=true
+            TTS_PROVIDER="ElevenLabs"
+        fi
+    fi
+    if grep -q "GOOGLE_CLOUD_API_KEY=\|GOOGLE_API_KEY=" "$ENV_FILE"; then
+        GC_KEY=$(grep -m1 "GOOGLE_CLOUD_API_KEY=\|GOOGLE_API_KEY=" "$ENV_FILE" | cut -d'=' -f2)
+        if [ -n "$GC_KEY" ]; then
+            echo -e "${GREEN}OK Google Cloud API key configured${NC}"
+            TTS_CONFIGURED=true
+            [ "$TTS_PROVIDER" = "none" ] && TTS_PROVIDER="Google Cloud"
+        fi
+    fi
+fi
+
+if [ "$TTS_CONFIGURED" = false ]; then
+    echo -e "${YELLOW}! No TTS API key found${NC}"
     echo
-    echo "To enable AI voices, add your ElevenLabs API key to ~/.env:"
-    echo "  echo 'ELEVENLABS_API_KEY=your_api_key_here' >> ~/.env"
-    echo "  Get a free key at: https://elevenlabs.io"
+    echo "To enable AI voices, add an API key to ~/.env:"
+    echo "  ElevenLabs: echo 'ELEVENLABS_API_KEY=your_key' >> ~/.env"
+    echo "  Google Cloud: echo 'GOOGLE_CLOUD_API_KEY=your_key' >> ~/.env"
     echo
 fi
 
-# Create LaunchAgent plist
-echo -e "${YELLOW}> Creating LaunchAgent configuration...${NC}"
-mkdir -p "$HOME/Library/LaunchAgents"
+# Create service configuration
+if [ "$OS_TYPE" = "Darwin" ]; then
+    echo -e "${YELLOW}> Creating LaunchAgent configuration...${NC}"
+    mkdir -p "$HOME/Library/LaunchAgents"
 
-cat > "$PLIST_PATH" << EOF
+    cat > "$PLIST_PATH" << EOF
 <?xml version="1.0" encoding="UTF-8"?>
 <!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
 <plist version="1.0">
@@ -123,15 +182,57 @@ cat > "$PLIST_PATH" << EOF
 </plist>
 EOF
 
-echo -e "${GREEN}OK LaunchAgent configuration created${NC}"
+    echo -e "${GREEN}OK LaunchAgent configuration created${NC}"
 
-# Load the LaunchAgent
-echo -e "${YELLOW}> Starting voice server service...${NC}"
-launchctl load "$PLIST_PATH" 2>/dev/null || {
-    echo -e "${RED}X Failed to load LaunchAgent${NC}"
-    echo "  Try manually: launchctl load $PLIST_PATH"
-    exit 1
-}
+    # Load the LaunchAgent
+    echo -e "${YELLOW}> Starting voice server service...${NC}"
+    launchctl load "$PLIST_PATH" 2>/dev/null || {
+        echo -e "${RED}X Failed to load LaunchAgent${NC}"
+        echo "  Try manually: launchctl load $PLIST_PATH"
+        exit 1
+    }
+
+elif [ "$OS_TYPE" = "Linux" ]; then
+    echo -e "${YELLOW}> Creating systemd user service...${NC}"
+    mkdir -p "$SYSTEMD_DIR"
+    mkdir -p "$(dirname "$LOG_PATH")"
+
+    # Resolve bun path for systemd
+    BUN_PATH="$(which bun)"
+
+    cat > "$SERVICE_PATH" << EOF
+[Unit]
+Description=PAI Voice Server
+After=network.target
+
+[Service]
+Type=simple
+ExecStart=${BUN_PATH} run ${SCRIPT_DIR}/server.ts
+WorkingDirectory=${SCRIPT_DIR}
+Restart=on-failure
+RestartSec=5
+Environment=HOME=${HOME}
+Environment=PATH=/usr/local/bin:/usr/bin:/bin:${HOME}/.bun/bin
+StandardOutput=append:${LOG_PATH}
+StandardError=append:${LOG_PATH}
+
+[Install]
+WantedBy=default.target
+EOF
+
+    echo -e "${GREEN}OK systemd service created${NC}"
+
+    # Enable and start the service
+    echo -e "${YELLOW}> Starting voice server service...${NC}"
+    systemctl --user daemon-reload
+    systemctl --user enable "$SERVICE_NAME" 2>/dev/null || true
+    systemctl --user start "$SERVICE_NAME" || {
+        echo -e "${RED}X Failed to start systemd service${NC}"
+        echo "  Try manually: systemctl --user start $SERVICE_NAME"
+        echo "  Check logs: journalctl --user -u $SERVICE_NAME"
+        exit 1
+    }
+fi
 
 # Wait for server to start
 sleep 2
@@ -161,15 +262,16 @@ echo -e "${GREEN}     Installation Complete!${NC}"
 echo -e "${GREEN}=====================================================${NC}"
 echo
 echo -e "${BLUE}Service Information:${NC}"
+echo "  - Platform: $OS_TYPE"
 echo "  - Service: $SERVICE_NAME"
 echo "  - Status: Running"
 echo "  - Port: 8888"
 echo "  - Logs: $LOG_PATH"
 
-if [ "$ELEVENLABS_CONFIGURED" = true ]; then
-    echo "  - Voice: ElevenLabs AI"
+if [ "$TTS_CONFIGURED" = true ]; then
+    echo "  - Voice: $TTS_PROVIDER"
 else
-    echo "  - Voice: macOS Say (fallback)"
+    echo "  - Voice: Not configured (add API key to ~/.env)"
 fi
 
 echo
@@ -180,6 +282,14 @@ echo "  - Start:    ./start.sh"
 echo "  - Restart:  ./restart.sh"
 echo "  - Uninstall: ./uninstall.sh"
 
+if [ "$OS_TYPE" = "Linux" ]; then
+    echo
+    echo -e "${BLUE}systemd Commands:${NC}"
+    echo "  - Status:  systemctl --user status $SERVICE_NAME"
+    echo "  - Logs:    journalctl --user -u $SERVICE_NAME -f"
+    echo "  - Restart: systemctl --user restart $SERVICE_NAME"
+fi
+
 echo
 echo -e "${BLUE}Test the server:${NC}"
 echo "  curl -X POST http://localhost:8888/notify \\"
@@ -189,18 +299,20 @@ echo "    -d '{\"message\": \"Hello from PAI\"}'"
 echo
 echo -e "${GREEN}The voice server will now start automatically when you log in.${NC}"
 
-# Ask about menu bar indicator
-echo
-read -p "Would you like to install a menu bar indicator? (y/n): " -n 1 -r
-echo
-if [[ $REPLY =~ ^[Yy]$ ]]; then
-    echo -e "${YELLOW}> Installing menu bar indicator...${NC}"
-    if [ -f "$SCRIPT_DIR/menubar/install-menubar.sh" ]; then
-        chmod +x "$SCRIPT_DIR/menubar/install-menubar.sh"
-        "$SCRIPT_DIR/menubar/install-menubar.sh"
-    else
-        echo -e "${YELLOW}! Menu bar installer not found${NC}"
-        echo "  You can install it manually later from:"
-        echo "  $SCRIPT_DIR/menubar/install-menubar.sh"
+# Ask about menu bar indicator (macOS only)
+if [ "$OS_TYPE" = "Darwin" ]; then
+    echo
+    read -p "Would you like to install a menu bar indicator? (y/n): " -n 1 -r
+    echo
+    if [[ $REPLY =~ ^[Yy]$ ]]; then
+        echo -e "${YELLOW}> Installing menu bar indicator...${NC}"
+        if [ -f "$SCRIPT_DIR/menubar/install-menubar.sh" ]; then
+            chmod +x "$SCRIPT_DIR/menubar/install-menubar.sh"
+            "$SCRIPT_DIR/menubar/install-menubar.sh"
+        else
+            echo -e "${YELLOW}! Menu bar installer not found${NC}"
+            echo "  You can install it manually later from:"
+            echo "  $SCRIPT_DIR/menubar/install-menubar.sh"
+        fi
     fi
 fi
diff --git a/Releases/v4.0.3/.claude/VoiceServer/server.ts b/Releases/v4.0.3/.claude/VoiceServer/server.ts
index 9f5dec95c..3be02aefe 100644
--- a/Releases/v4.0.3/.claude/VoiceServer/server.ts
+++ b/Releases/v4.0.3/.claude/VoiceServer/server.ts
@@ -1,6 +1,13 @@
 #!/usr/bin/env bun
 /**
- * Voice Server - Personal AI Voice notification server using ElevenLabs TTS
+ * Voice Server - Personal AI Voice notification server with multi-provider TTS
+ *
+ * Supported TTS providers:
+ *   - ElevenLabs (default) — high-quality AI voices, 10K free chars/month
+ *   - Google Cloud TTS — WaveNet/Neural2/Standard voices, up to 4M free chars/month
+ *
+ * Provider selection: settings.json daidentity.ttsProvider ("elevenlabs" | "google-cloud")
+ * Falls back to ElevenLabs if not specified (backwards compatible).
  *
  * Architecture: Pure pass-through. All voice config comes from settings.json.
  * The server has zero hardcoded voice parameters.
@@ -11,7 +18,7 @@
  *   3. Neither → use settings.json daidentity.voices.main as default
  *
  * Pronunciation preprocessing: loads pronunciations.json and applies
- * word-boundary replacements before sending text to ElevenLabs TTS.
+ * word-boundary replacements before sending text to TTS.
  */
 
 import { serve } from "bun";
@@ -34,11 +41,9 @@ if (existsSync(envPath)) {
 
 const PORT = parseInt(process.env.PORT || "8888");
 const ELEVENLABS_API_KEY = process.env.ELEVENLABS_API_KEY;
+const GOOGLE_CLOUD_API_KEY = process.env.GOOGLE_CLOUD_API_KEY || process.env.GOOGLE_API_KEY;
 
-if (!ELEVENLABS_API_KEY) {
-  console.error('⚠️  ELEVENLABS_API_KEY not found in ~/.env');
-  console.error('Add: ELEVENLABS_API_KEY=your_key_here');
-}
+// TTS provider is resolved after settings.json is loaded (see below)
 
 // ==========================================================================
 // Pronunciation System
@@ -130,9 +135,31 @@ interface VoiceEntry {
   volume: number;
 }
 
+// TTS provider type
+type TtsProvider = 'elevenlabs' | 'google-cloud';
+
+// Google Cloud TTS voice configuration
+interface GoogleCloudVoiceConfig {
+  languageCode: string;   // e.g. "en-US"
+  voiceName: string;      // e.g. "en-US-Neural2-D"
+  voiceType: 'STANDARD' | 'WAVENET' | 'NEURAL2';
+  speakingRate: number;   // 0.25 to 4.0, default 1.0
+  pitch: number;          // -20.0 to 20.0, default 0.0
+}
+
+const FALLBACK_GOOGLE_VOICE: GoogleCloudVoiceConfig = {
+  languageCode: 'en-US',
+  voiceName: 'en-US-Neural2-D',
+  voiceType: 'NEURAL2',
+  speakingRate: 1.0,
+  pitch: 0.0,
+};
+
 // Loaded config from settings.json
 interface LoadedVoiceConfig {
   defaultVoiceId: string;
+  ttsProvider: TtsProvider;
+  googleVoice: GoogleCloudVoiceConfig;
   voices: Record<string, VoiceEntry>;     // keyed by name ("main", "algorithm")
   voicesByVoiceId: Record<string, VoiceEntry>;  // keyed by voiceId for lookup
   desktopNotifications: boolean;  // whether to show macOS notification banners
@@ -155,7 +182,7 @@ function loadVoiceConfig(): LoadedVoiceConfig {
   try {
     if (!existsSync(settingsPath)) {
       console.warn('⚠️  settings.json not found — using fallback voice defaults');
-      return { defaultVoiceId: '', voices: {}, voicesByVoiceId: {}, desktopNotifications: true };
+      return { defaultVoiceId: '', ttsProvider: 'elevenlabs', googleVoice: FALLBACK_GOOGLE_VOICE, voices: {}, voicesByVoiceId: {}, desktopNotifications: true };
     }
 
     const content = readFileSync(settingsPath, 'utf-8');
@@ -189,16 +216,30 @@ function loadVoiceConfig(): LoadedVoiceConfig {
     // Default voice ID from settings
     const defaultVoiceId = voices.main?.voiceId || daidentity.mainDAVoiceID || '';
 
+    // TTS provider selection
+    const ttsProvider: TtsProvider = daidentity.ttsProvider === 'google-cloud' ? 'google-cloud' : 'elevenlabs';
+
+    // Google Cloud voice config
+    const gcVoice = daidentity.googleCloudVoice || {};
+    const googleVoice: GoogleCloudVoiceConfig = {
+      languageCode: gcVoice.languageCode || FALLBACK_GOOGLE_VOICE.languageCode,
+      voiceName: gcVoice.voiceName || FALLBACK_GOOGLE_VOICE.voiceName,
+      voiceType: gcVoice.voiceType || FALLBACK_GOOGLE_VOICE.voiceType,
+      speakingRate: gcVoice.speakingRate ?? FALLBACK_GOOGLE_VOICE.speakingRate,
+      pitch: gcVoice.pitch ?? FALLBACK_GOOGLE_VOICE.pitch,
+    };
+
     const voiceNames = Object.keys(voices);
     console.log(`✅ Loaded ${voiceNames.length} voice config(s) from settings.json: ${voiceNames.join(', ')}`);
+    console.log(`🔊 TTS provider: ${ttsProvider}`);
     for (const [name, entry] of Object.entries(voices)) {
       console.log(`   ${name}: ${entry.voiceName || entry.voiceId} (speed: ${entry.speed}, stability: ${entry.stability})`);
     }
 
-    return { defaultVoiceId, voices, voicesByVoiceId, desktopNotifications };
+    return { defaultVoiceId, ttsProvider, googleVoice, voices, voicesByVoiceId, desktopNotifications };
   } catch (error) {
     console.error('⚠️  Failed to load settings.json voice config:', error);
-    return { defaultVoiceId: '', voices: {}, voicesByVoiceId: {}, desktopNotifications: true };
+    return { defaultVoiceId: '', ttsProvider: 'elevenlabs', googleVoice: FALLBACK_GOOGLE_VOICE, voices: {}, voicesByVoiceId: {}, desktopNotifications: true };
   }
 }
 
@@ -330,19 +371,13 @@ function validateInput(input: any): { valid: boolean; error?: string; sanitized?
 }
 
 // Generate speech using ElevenLabs API — pure pass-through of voice_settings
-async function generateSpeech(
+async function generateSpeechElevenLabs(
   text: string,
   voiceId: string,
   voiceSettings: ElevenLabsVoiceSettings
 ): Promise<ArrayBuffer> {
   if (!ELEVENLABS_API_KEY) {
-    throw new Error('ElevenLabs API key not configured');
-  }
-
-  // Apply pronunciation replacements before sending to TTS
-  const pronouncedText = applyPronunciations(text);
-  if (pronouncedText !== text) {
-    console.log(`📖 Pronunciation: "${text}" → "${pronouncedText}"`);
+    throw new Error('ElevenLabs API key not configured — add ELEVENLABS_API_KEY to ~/.env');
   }
 
   const url = `https://api.elevenlabs.io/v1/text-to-speech/${voiceId}`;
@@ -355,7 +390,7 @@ async function generateSpeech(
       'xi-api-key': ELEVENLABS_API_KEY,
     },
     body: JSON.stringify({
-      text: pronouncedText,
+      text,
       model_id: 'eleven_turbo_v2_5',
       voice_settings: voiceSettings,
     }),
@@ -369,14 +404,110 @@ async function generateSpeech(
   return await response.arrayBuffer();
 }
 
-// Play audio using afplay (macOS)
+// Generate speech using Google Cloud Text-to-Speech REST API (no SDK dependency)
+async function generateSpeechGoogleCloud(
+  text: string,
+  gcVoice: GoogleCloudVoiceConfig
+): Promise<ArrayBuffer> {
+  if (!GOOGLE_CLOUD_API_KEY) {
+    throw new Error('Google Cloud API key not configured — add GOOGLE_CLOUD_API_KEY to ~/.env');
+  }
+
+  const url = `https://texttospeech.googleapis.com/v1/text:synthesize?key=${GOOGLE_CLOUD_API_KEY}`;
+
+  const response = await fetch(url, {
+    method: 'POST',
+    headers: { 'Content-Type': 'application/json' },
+    body: JSON.stringify({
+      input: { text },
+      voice: {
+        languageCode: gcVoice.languageCode,
+        name: gcVoice.voiceName,
+      },
+      audioConfig: {
+        audioEncoding: 'MP3',
+        speakingRate: gcVoice.speakingRate,
+        pitch: gcVoice.pitch,
+      },
+    }),
+  });
+
+  if (!response.ok) {
+    const errorText = await response.text();
+    throw new Error(`Google Cloud TTS API error: ${response.status} - ${errorText}`);
+  }
+
+  const data = await response.json() as { audioContent: string };
+
+  // Google returns base64-encoded audio — decode to ArrayBuffer
+  const binaryString = atob(data.audioContent);
+  const bytes = new Uint8Array(binaryString.length);
+  for (let i = 0; i < binaryString.length; i++) {
+    bytes[i] = binaryString.charCodeAt(i);
+  }
+  return bytes.buffer;
+}
+
+// Route TTS generation to the configured provider
+async function generateSpeech(
+  text: string,
+  voiceId: string,
+  voiceSettings: ElevenLabsVoiceSettings
+): Promise<ArrayBuffer> {
+  // Apply pronunciation replacements before sending to TTS
+  const pronouncedText = applyPronunciations(text);
+  if (pronouncedText !== text) {
+    console.log(`📖 Pronunciation: "${text}" → "${pronouncedText}"`);
+  }
+
+  if (voiceConfig.ttsProvider === 'google-cloud') {
+    return generateSpeechGoogleCloud(pronouncedText, voiceConfig.googleVoice);
+  }
+  return generateSpeechElevenLabs(pronouncedText, voiceId, voiceSettings);
+}
+
+// Detect platform and pick audio player
+function getAudioPlayer(): { command: string; args: (file: string, volume: number) => string[] } {
+  const platform = process.platform;
+
+  if (platform === 'darwin') {
+    return {
+      command: '/usr/bin/afplay',
+      args: (file, volume) => ['-v', volume.toString(), file],
+    };
+  }
+
+  // Linux: try mpv, ffplay, aplay in order
+  const linuxPlayers: Array<{ cmd: string; args: (file: string, volume: number) => string[] }> = [
+    { cmd: 'mpv', args: (file, volume) => ['--no-video', `--volume=${Math.round(volume * 100)}`, file] },
+    { cmd: 'ffplay', args: (file, _volume) => ['-nodisp', '-autoexit', '-loglevel', 'quiet', file] },
+    { cmd: 'aplay', args: (file, _volume) => [file] },
+  ];
+
+  for (const player of linuxPlayers) {
+    try {
+      const result = Bun.spawnSync(['which', player.cmd]);
+      if (result.exitCode === 0) {
+        return { command: player.cmd, args: player.args };
+      }
+    } catch { /* try next */ }
+  }
+
+  // Fallback to mpv (will error at runtime if not installed)
+  return { command: 'mpv', args: (file, volume) => ['--no-video', `--volume=${Math.round(volume * 100)}`, file] };
+}
+
+const audioPlayer = getAudioPlayer();
+console.log(`🔈 Audio player: ${audioPlayer.command} (${process.platform})`);
+
+// Play audio using platform-appropriate player
 async function playAudio(audioBuffer: ArrayBuffer, volume: number = FALLBACK_VOLUME): Promise<void> {
   const tempFile = `/tmp/voice-${Date.now()}.mp3`;
 
   await Bun.write(tempFile, audioBuffer);
 
   return new Promise((resolve, reject) => {
-    const proc = spawn('/usr/bin/afplay', ['-v', volume.toString(), tempFile]);
+    const proc = spawn(audioPlayer.command, audioPlayer.args(tempFile, volume));
 
     proc.on('error', (error) => {
       console.error('Error playing audio:', error);
@@ -388,7 +519,7 @@ async function playAudio(audioBuffer: ArrayBuffer, volume: number = FALLBACK_VOL
       if (code === 0) {
         resolve();
       } else {
-        reject(new Error(`afplay exited with code ${code}`));
+        reject(new Error(`${audioPlayer.command} exited with code ${code}`));
       }
     });
   });
@@ -512,13 +643,20 @@ async function sendNotification(
     }
   }
 
-  // Display macOS notification (can be disabled via settings.json: notifications.desktop.enabled: false)
+  // Display desktop notification (can be disabled via settings.json: notifications.desktop.enabled: false)
   if (voiceConfig.desktopNotifications) {
     try {
-      const escapedTitle = escapeForAppleScript(safeTitle);
-      const escapedMessage = escapeForAppleScript(safeMessage);
-      const script = `display notification "${escapedMessage}" with title "${escapedTitle}" sound name ""`;
-      await spawnSafe('/usr/bin/osascript', ['-e', script]);
+      if (process.platform === 'darwin') {
+        const escapedTitle = escapeForAppleScript(safeTitle);
+        const escapedMessage = escapeForAppleScript(safeMessage);
+        const script = `display notification "${escapedMessage}" with title "${escapedTitle}" sound name ""`;
+        await spawnSafe('/usr/bin/osascript', ['-e', script]);
+      } else {
+        // Linux: try notify-send if available
+        try {
+          await spawnSafe('notify-send', [safeTitle, safeMessage]);
+        } catch { /* notify-send not available, skip silently */ }
+      }
     } catch (error) {
       console.error("Notification display error:", error);
     }
@@ -688,9 +826,11 @@ const server = serve({
         JSON.stringify({
           status: "healthy",
           port: PORT,
-          voice_system: "ElevenLabs",
+          voice_system: voiceConfig.ttsProvider,
           default_voice_id: DEFAULT_VOICE_ID,
-          api_key_configured: !!ELEVENLABS_API_KEY,
+          elevenlabs_configured: !!ELEVENLABS_API_KEY,
+          google_cloud_configured: !!GOOGLE_CLOUD_API_KEY,
+          audio_player: audioPlayer.command,
           pronunciation_rules: pronunciationRules.length,
           configured_voices: Object.keys(voiceConfig.voices),
         }),
@@ -709,8 +849,14 @@ const server = serve({
 });
 
 console.log(`🚀 Voice Server running on port ${PORT}`);
-console.log(`🎙️  Using ElevenLabs TTS (default voice: ${DEFAULT_VOICE_ID})`);
+console.log(`🔊 TTS provider: ${voiceConfig.ttsProvider}`);
+if (voiceConfig.ttsProvider === 'google-cloud') {
+  console.log(`🎙️  Using Google Cloud TTS (voice: ${voiceConfig.googleVoice.voiceName}, type: ${voiceConfig.googleVoice.voiceType})`);
+  console.log(`🔑 Google Cloud API Key: ${GOOGLE_CLOUD_API_KEY ? '✅ Configured' : '❌ Missing'}`);
+} else {
+  console.log(`🎙️  Using ElevenLabs TTS (default voice: ${DEFAULT_VOICE_ID})`);
+  console.log(`🔑 ElevenLabs API Key: ${ELEVENLABS_API_KEY ? '✅ Configured' : '❌ Missing'}`);
+}
 console.log(`📡 POST to http://localhost:${PORT}/notify`);
 console.log(`🔒 Security: CORS restricted to localhost, rate limiting enabled`);
-console.log(`🔑 API Key: ${ELEVENLABS_API_KEY ? '✅ Configured' : '❌ Missing'}`);
 console.log(`📖 Pronunciations: ${pronunciationRules.length} rules loaded`);