diff --git a/src/aiy/cloudspeech.py b/src/aiy/cloudspeech.py
index 184ed9b5..160fa63f 100644
--- a/src/aiy/cloudspeech.py
+++ b/src/aiy/cloudspeech.py
@@ -36,18 +36,30 @@ def __init__(self, credentials_file):
         self._recorder = aiy.audio.get_recorder()
         self._hotwords = []
 
-    def recognize(self):
+    def recognize(self, immediate=False):
         """Recognizes the user's speech and transcript it into text.
 
         This function listens to the user's speech via the VoiceHat speaker. Then it
         contacts Google CloudSpeech APIs and returns a textual transcript if possible.
         If hotword list is populated this method will only respond if hotword is said.
+
+        Args:
+            immediate: ignore the hotword list, even if it has been populated
+                may be used to create a conversational experience, for example:
+
+                text = recognizer.recognize()
+                if 'call a friend' in text:
+                    aiy.audio.say('OK, which one?')
+                    friend = recognizer.recognize(immediate=True)
+                    make_a_call(friend)
         """
         self._request.reset()
         self._request.set_endpointer_cb(self._endpointer_callback)
         self._recorder.add_processor(self._request)
         text = self._request.do_request().transcript
-        if self._hotwords and text:
+        if immediate:
+            return text
+        elif self._hotwords and text:
             text = text.lower()
             loc_min = len(text)
             hotword_found = ''