Gender option

app.py

@@ -29,20 +29,7 @@ else:
 tts = TTS(model_name, gpu=torch.cuda.is_available())
 tts.to(device_type)
 
-def predict(prompt, language, audio_file_pth, mic_file_path, use_mic):
-    if use_mic:
-        if mic_file_path is None:
-            gr.Warning("Please record your voice with Microphone, or uncheck Use Microphone to use reference audios")
-            return (
-                None,
-                None,
-                None,
-            )
-        else:
-            speaker_wav = mic_file_path
-    else:
-        speaker_wav = audio_file_pth
-
+def predict(prompt, language, gender, audio_file_pth, mic_file_path, use_mic):
     if len(prompt) < 2:
         gr.Warning("Please give a longer prompt text")
         return (
@@ -56,7 +43,27 @@ def predict(prompt, language, audio_file_pth, mic_file_path, use_mic):
                 None,
                 None,
                 None,
-            )  
+            )
+
+    if use_mic:
+        if mic_file_path is None:
+            gr.Warning("Please record your voice with Microphone, or uncheck Use Microphone to use reference audios")
+            return (
+                None,
+                None,
+                None,
+            )
+        else:
+            speaker_wav = mic_file_path
+    else:
+        speaker_wav = audio_file_pth
+
+    if speaker_wav is None:
+        if gender == "male":
+            speaker_wav = "examples/male.wav"
+        else:
+            speaker_wav = "examples/female.wav"
+        
     try:
         if language == "fr":
             if m.find("your") != -1:
@@ -136,11 +143,12 @@ Leave a star on the Github <a href="https://github.com/coqui-ai/TTS">TTS</a>, wh
             max_choices=1,
             value="en",
         )
+        gender = gr.Radio(["female", "male"], label="Gender", info="Gender of the voice")
         audio_file_pth = gr.Audio(
             label="Reference Audio",
             #info="Click on the ✎ button to upload your own target speaker audio",
             type="filepath",
-            value="examples/female.wav",
+            value=None,
         )
         mic_file_path = gr.Audio(sources=["microphone"],
                  type="filepath",
@@ -159,7 +167,7 @@ Leave a star on the Github <a href="https://github.com/coqui-ai/TTS">TTS</a>, wh
         information = gr.HTML()
 
     submit.click(predict, inputs = [
-        prompt, language, audio_file_pth, mic_file_path, use_mic
+        prompt, language, gender, audio_file_pth, mic_file_path, use_mic
     ], outputs = [
         waveform_visual,
         synthesised_audio,