diff --git a/api/core/tools/provider/builtin/azurespeech/tools/asr.py b/api/core/tools/provider/builtin/azurespeech/tools/asr.py
index a3b1a27a312f16..3f6cf2b34b005f 100644
--- a/api/core/tools/provider/builtin/azurespeech/tools/asr.py
+++ b/api/core/tools/provider/builtin/azurespeech/tools/asr.py
@@ -19,7 +19,7 @@ def _invoke(self, user_id: str, tool_parameters: dict[str, Any]) -> list[ToolInv
 
         files: dict = {"audio": audio_binary}
         if definition:
-            files["definition"] = definition
+            files["definition"] = '"{}"'.format(definition)
         resp = requests.post(
             "https://{}.api.cognitive.microsoft.com/speechtotext/transcriptions:transcribe?api-version={}".format(
                 self.runtime.credentials.get("azure_speech_region"),