diff --git a/CHANGELOG.md b/CHANGELOG.md
index 4b4c4940022..04f0af64a26 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -41,6 +41,8 @@ Docs: https://docs.openclaw.ai
 - Browser/CDP: honor configured remote and `attachOnly` CDP HTTP/WebSocket
   timeouts when opening tabs through raw CDP or `/json/new` fallback. (#54238)
   Thanks @FuncWei.
+- WhatsApp/TTS: send visible text separately from PTT voice-note audio instead
+  of relying on hidden voice-note captions. Fixes #51081.
 - Agents/TTS: preserve `[[audio_as_voice]]` directives on trusted text
   tool-result `MEDIA:` payloads so generated audio still delivers as a voice
   note. (#46535) Thanks @azade-c.
diff --git a/docs/channels/whatsapp.md b/docs/channels/whatsapp.md
index 74dfd52677e..239904d1ea5 100644
--- a/docs/channels/whatsapp.md
+++ b/docs/channels/whatsapp.md
@@ -365,7 +365,7 @@ When the linked self number is also present in `allowFrom`, WhatsApp self-chat s
     - non-Ogg audio, including Microsoft Edge TTS MP3/WebM output, is transcoded to Ogg/Opus before PTT delivery
     - native Ogg/Opus audio is sent with `audio/ogg; codecs=opus` for voice-note compatibility
     - animated GIF playback is supported via `gifPlayback: true` on video sends
-    - captions are applied to the first media item when sending multi-media reply payloads
+    - captions are applied to the first media item when sending multi-media reply payloads, except PTT voice notes send the audio first and visible text separately because WhatsApp clients do not render voice-note captions consistently
     - media source can be HTTP(S), `file://`, or local paths
   </Accordion>
 
diff --git a/docs/tools/tts.md b/docs/tools/tts.md
index 6012df221b3..0af905c65f8 100644
--- a/docs/tools/tts.md
+++ b/docs/tools/tts.md
@@ -664,6 +664,8 @@ reply delivery. When the channel is Feishu, Matrix, Telegram, or WhatsApp,
 the audio is delivered as a voice message rather than a file attachment.
 Feishu can transcode non-Opus TTS output on this path when `ffmpeg` is
 available.
+WhatsApp sends visible text separately from PTT voice-note audio because clients
+do not consistently render captions on voice notes.
 It accepts optional `channel` and `timeoutMs` fields; `timeoutMs` is a
 per-call provider request timeout in milliseconds.
 
diff --git a/extensions/whatsapp/src/auto-reply/deliver-reply.test.ts b/extensions/whatsapp/src/auto-reply/deliver-reply.test.ts
index 9124ba1869e..5bb4c62f2bf 100644
--- a/extensions/whatsapp/src/auto-reply/deliver-reply.test.ts
+++ b/extensions/whatsapp/src/auto-reply/deliver-reply.test.ts
@@ -91,6 +91,12 @@ function mockFirstReplyFailureWithWrappedError(msg: WebInboundMsg, message: stri
   });
 }
 
+function expectFirstSendMediaPayload(msg: WebInboundMsg) {
+  const payload = vi.mocked(msg.sendMedia).mock.calls[0]?.[0];
+  expect(payload).toBeDefined();
+  return payload;
+}
+
 function mockSecondReplySuccess(msg: WebInboundMsg) {
   (msg.reply as unknown as { mockResolvedValueOnce: (v: unknown) => void }).mockResolvedValueOnce(
     undefined,
@@ -524,14 +530,14 @@ describe("deliverWebReply", () => {
         audio: expect.any(Buffer),
         ptt: true,
         mimetype: "audio/ogg; codecs=opus",
-        caption: "caption",
       }),
       undefined,
     );
-    expect(msg.reply).not.toHaveBeenCalled();
+    expect(expectFirstSendMediaPayload(msg)).not.toHaveProperty("caption");
+    expect(msg.reply).toHaveBeenCalledWith("caption", undefined);
   });
 
-  it("sends audio media as ptt voice note", async () => {
+  it("sends audio media as ptt voice note with visible text separately", async () => {
     const msg = makeMsg();
     (
       loadWebMedia as unknown as { mockResolvedValueOnce: (v: unknown) => void }
@@ -555,10 +561,11 @@ describe("deliverWebReply", () => {
         audio: expect.any(Buffer),
         ptt: true,
         mimetype: "audio/ogg; codecs=opus",
-        caption: "cap",
       }),
       undefined,
     );
+    expect(expectFirstSendMediaPayload(msg)).not.toHaveProperty("caption");
+    expect(msg.reply).toHaveBeenCalledWith("cap", undefined);
   });
 
   it("transcodes mp3 audio media before sending a ptt voice note", async () => {
@@ -594,10 +601,11 @@ describe("deliverWebReply", () => {
         audio: Buffer.from("opus-output"),
         ptt: true,
         mimetype: "audio/ogg; codecs=opus",
-        caption: "cap",
       }),
       undefined,
     );
+    expect(expectFirstSendMediaPayload(msg)).not.toHaveProperty("caption");
+    expect(msg.reply).toHaveBeenCalledWith("cap", undefined);
   });
 
   it("sends video media", async () => {
diff --git a/extensions/whatsapp/src/auto-reply/deliver-reply.ts b/extensions/whatsapp/src/auto-reply/deliver-reply.ts
index cfd16737df4..03768490536 100644
--- a/extensions/whatsapp/src/auto-reply/deliver-reply.ts
+++ b/extensions/whatsapp/src/auto-reply/deliver-reply.ts
@@ -156,12 +156,14 @@ export async function deliverWebReply(params: {
                 audio: media.buffer,
                 ptt: true,
                 mimetype: media.mimetype,
-                caption,
               },
               quote,
             ),
           "media:audio",
         );
+        if (caption) {
+          await sendWithRetry(() => msg.reply(caption, quote), "media:audio-text");
+        }
       } else if (media.kind === "video") {
         const quote = getQuote();
         await sendWithRetry(
diff --git a/extensions/whatsapp/src/inbound/send-api.test.ts b/extensions/whatsapp/src/inbound/send-api.test.ts
index 7dec794a14c..b76779bc78b 100644
--- a/extensions/whatsapp/src/inbound/send-api.test.ts
+++ b/extensions/whatsapp/src/inbound/send-api.test.ts
@@ -100,6 +100,23 @@ describe("createWebSendApi", () => {
     });
   });
 
+  it("sends visible text separately from push-to-talk voice notes", async () => {
+    const payload = Buffer.from("aud");
+    await api.sendMessage("+1555", "voice text", payload, "audio/ogg");
+    expect(sendMessage).toHaveBeenNthCalledWith(
+      1,
+      "1555@s.whatsapp.net",
+      expect.objectContaining({
+        audio: payload,
+        ptt: true,
+        mimetype: "audio/ogg",
+      }),
+    );
+    expect(sendMessage).toHaveBeenNthCalledWith(2, "1555@s.whatsapp.net", {
+      text: "voice text",
+    });
+  });
+
   it("supports video media and gifPlayback option", async () => {
     const payload = Buffer.from("vid");
     await api.sendMessage("+1555", "cap", payload, "video/mp4", { gifPlayback: true });
diff --git a/extensions/whatsapp/src/inbound/send-api.ts b/extensions/whatsapp/src/inbound/send-api.ts
index 92ccef288ba..69aefcf9f8e 100644
--- a/extensions/whatsapp/src/inbound/send-api.ts
+++ b/extensions/whatsapp/src/inbound/send-api.ts
@@ -85,6 +85,14 @@ export function createWebSendApi(params: {
       const result = quotedOpts
         ? await params.sock.sendMessage(jid, payload, quotedOpts)
         : await params.sock.sendMessage(jid, payload);
+      if (mediaBuffer && mediaType?.startsWith("audio/") && text.trim()) {
+        const textPayload: AnyMessageContent = { text };
+        if (quotedOpts) {
+          await params.sock.sendMessage(jid, textPayload, quotedOpts);
+        } else {
+          await params.sock.sendMessage(jid, textPayload);
+        }
+      }
       const accountId = sendOptions?.accountId ?? params.defaultAccountId;
       recordWhatsAppOutbound(accountId);
       const messageId = resolveOutboundMessageId(result);
diff --git a/extensions/whatsapp/src/send.test.ts b/extensions/whatsapp/src/send.test.ts
index 9d96bd9860d..0ccd2df1f50 100644
--- a/extensions/whatsapp/src/send.test.ts
+++ b/extensions/whatsapp/src/send.test.ts
@@ -245,12 +245,8 @@ describe("web outbound", () => {
       cfg: WHATSAPP_TEST_CFG,
       mediaUrl: "/tmp/voice.ogg",
     });
-    expect(sendMessage).toHaveBeenLastCalledWith(
-      "+1555",
-      "voice note",
-      buf,
-      "audio/ogg; codecs=opus",
-    );
+    expect(sendMessage).toHaveBeenNthCalledWith(1, "+1555", "", buf, "audio/ogg; codecs=opus");
+    expect(sendMessage).toHaveBeenNthCalledWith(2, "+1555", "voice note", undefined, undefined);
   });
 
   it.each([
@@ -274,12 +270,14 @@ describe("web outbound", () => {
     expect(hoisted.runFfmpeg).toHaveBeenCalledWith(
       expect.arrayContaining(["-c:a", "libopus", "-ar", "48000", "-b:a", "64k"]),
     );
-    expect(sendMessage).toHaveBeenLastCalledWith(
+    expect(sendMessage).toHaveBeenNthCalledWith(
+      1,
       "+1555",
-      "voice note",
+      "",
       Buffer.from("opus-output"),
       "audio/ogg; codecs=opus",
     );
+    expect(sendMessage).toHaveBeenNthCalledWith(2, "+1555", "voice note", undefined, undefined);
   });
 
   it("maps video with caption", async () => {
diff --git a/extensions/whatsapp/src/send.ts b/extensions/whatsapp/src/send.ts
index 4a54c82a6b1..da698c09b26 100644
--- a/extensions/whatsapp/src/send.ts
+++ b/extensions/whatsapp/src/send.ts
@@ -115,6 +115,7 @@ export async function sendMessageWhatsApp(
     let mediaBuffer: Buffer | undefined;
     let mediaType: string | undefined;
     let documentFileName: string | undefined;
+    let visibleTextAfterVoice: string | undefined;
     if (primaryMediaUrl) {
       const media = await prepareWhatsAppOutboundMedia(
         await loadOutboundMediaFromUrl(primaryMediaUrl, {
@@ -128,7 +129,10 @@ export async function sendMessageWhatsApp(
       const caption = text || undefined;
       mediaBuffer = media.buffer;
       mediaType = media.mimetype;
-      if (media.kind === "document") {
+      if (media.kind === "audio" && caption) {
+        visibleTextAfterVoice = caption;
+        text = "";
+      } else if (media.kind === "document") {
         text = caption ?? "";
         documentFileName = media.fileName;
       } else {
@@ -152,6 +156,13 @@ export async function sendMessageWhatsApp(
     const result = sendOptions
       ? await active.sendMessage(to, text, mediaBuffer, mediaType, sendOptions)
       : await active.sendMessage(to, text, mediaBuffer, mediaType);
+    if (visibleTextAfterVoice) {
+      if (sendOptions) {
+        await active.sendMessage(to, visibleTextAfterVoice, undefined, undefined, sendOptions);
+      } else {
+        await active.sendMessage(to, visibleTextAfterVoice, undefined, undefined);
+      }
+    }
     const messageId = (result as { messageId?: string })?.messageId ?? "unknown";
     const durationMs = Date.now() - startedAt;
     outboundLog.info(