STT: Improve example, add fields for partials and errors

examples/speech/main.py

@@ -7,21 +7,30 @@
 from plyer import speech
 
 Builder.load_string('''
+#:import speech plyer.speech
+
 <SpeechInterface>:
     orientation: 'vertical'
     Label:
         size_hint_y: None
         height: sp(40)
-        text: 'Is supported: %s' % root.speech.exist()
+        text: 'Is supported: %s' % speech.exist()
     Label:
         size_hint_y: None
         height: sp(40)
         text: 'Possible Matches'
-    Label:
+    TextInput:
         id: results
-        size_hint_y: 0.7
-        height: sp(40)
-        text: ''
+        hint_text: 'results (auto stop)'
+        size_hint_y: 0.3
+    TextInput:
+        id: partial
+        hint_text: 'partial results (manual stop)'
+        size_hint_y: 0.3
+    TextInput:
+        id: errors
+        size_hint_y: None
+        height: sp(20)
     Button:
         id: start_button
         text: 'Start Listening'
@@ -34,41 +43,38 @@
 class SpeechInterface(BoxLayout):
     '''Root Widget.'''
 
-    speech = speech
-    state = StringProperty()
-
     def start_listening(self):
-        if self.speech.state == 'listening':
+        if speech.listening:
             self.stop_listening()
             return
 
-        start_button = self.ids['start_button']
+        start_button = self.ids.start_button
         start_button.text = 'Stop'
 
-        label_results = self.ids['results']
-        label_results.text = ''
+        self.ids.results.text = ''
+        self.ids.partial.text = ''
 
-        self.speech.start()
-        self.state = self.speech.state
+        speech.start()
 
         Clock.schedule_interval(self.check_state, 1/5)
 
     def stop_listening(self):
-        start_button = self.ids['start_button']
+        start_button = self.ids.start_button
         start_button.text = 'Start Listening'
 
-        self.speech.stop()
+        speech.stop()
         self.update()
 
         Clock.unschedule(self.check_state)
 
     def check_state(self, dt):
-        if self.state != self.speech.state:
+        # if the recognizer service stops, change UI
+        if not speech.listening:
             self.stop_listening()
 
     def update(self):
-        label_results = self.ids['results']
-        label_results.text = '\n'.join(set(self.speech.results))
+        self.ids.partial.text = '\n'.join(speech.partial_results)
+        self.ids.results.text = '\n'.join(speech.results)
 
 
 class SpeechApp(App):