fix(media): remap AAC uploads to M4A (#66446)

* fix(media): remap AAC uploads to M4A

* fix(media): remap AAC uploads to M4A

CHANGELOG.md

@@ -30,6 +30,7 @@ Docs: https://docs.openclaw.ai
 - Agents/context engine: compact engine-owned sessions from the first tool-loop delta and preserve ingest fallback when `afterTurn` is absent, so long-running tool loops can stay bounded without dropping engine state. (#63555) Thanks @Bikkies.
 - Discord/native commands: return the real status card for native `/status` interactions instead of falling through to the synthetic `✅ Done.` ack when the generic dispatcher produces no visible reply. (#54629) Thanks @tkozzer and @vincentkoc.
 - Hooks/Ollama: let LLM-backed session-memory slug generation honor an explicit `agents.defaults.timeoutSeconds` override instead of always aborting after 15 seconds, so slow local Ollama runs stop silently dropping back to generic filenames. (#66237) Thanks @dmak and @vincentkoc.
+- Media/transcription: remap `.aac` filenames to `.m4a` for OpenAI-compatible audio uploads so AAC voice notes stop failing MIME-sensitive transcription endpoints. (#66446) Thanks @ben-z.
 
 ## 2026.4.14-beta.1
 

src/media-understanding/openai-compatible-audio.test.ts

@@ -48,4 +48,26 @@ describe("transcribeOpenAiCompatibleAudio", () => {
     expect(headers.get("version")).toBeNull();
     expect(headers.get("user-agent")).toBeNull();
   });
+
+  it("remaps AAC uploads to an M4A filename before submitting the form", async () => {
+    const { fetchFn, getRequest } = createRequestCaptureJsonFetch({ text: "ok" });
+
+    await transcribeOpenAiCompatibleAudio({
+      buffer: Buffer.from("audio"),
+      fileName: "voice-note.aac",
+      mime: "audio/aac",
+      apiKey: "test-key",
+      timeoutMs: 1000,
+      fetchFn,
+      provider: "openai",
+      defaultBaseUrl: "https://api.openai.com/v1",
+      defaultModel: "gpt-4o-transcribe",
+    });
+
+    const form = getRequest().init?.body;
+    expect(form).toBeInstanceOf(FormData);
+    const file = (form as FormData).get("file");
+    expect(file).toBeInstanceOf(File);
+    expect((file as File).name).toBe("voice-note.m4a");
+  });
 });

src/media-understanding/openai-compatible-audio.ts

@@ -18,6 +18,20 @@ function resolveModel(model: string | undefined, fallback: string): string {
   return trimmed || fallback;
 }
 
+function resolveUploadFileName(fileName?: string, mime?: string): string {
+  const trimmed = fileName?.trim();
+  const baseName = trimmed ? path.basename(trimmed) : "audio";
+  const lowerMime = mime?.trim().toLowerCase();
+
+  if (/\.aac$/i.test(baseName)) {
+    return `${baseName.slice(0, -4) || "audio"}.m4a`;
+  }
+  if (!path.extname(baseName) && lowerMime === "audio/aac") {
+    return `${baseName || "audio"}.m4a`;
+  }
+  return baseName;
+}
+
 export async function transcribeOpenAiCompatibleAudio(
   params: OpenAiCompatibleAudioParams,
 ): Promise<AudioTranscriptionResult> {
@@ -40,7 +54,7 @@ export async function transcribeOpenAiCompatibleAudio(
 
   const model = resolveModel(params.model, params.defaultModel);
   const form = new FormData();
-  const fileName = params.fileName?.trim() || path.basename(params.fileName) || "audio";
+  const fileName = resolveUploadFileName(params.fileName, params.mime);
   const bytes = new Uint8Array(params.buffer);
   const blob = new Blob([bytes], {
     type: params.mime ?? "application/octet-stream",