Add some more sample voice

.vscode/settings.json

@@ -0,0 +1,3 @@
+{
+  "cSpell.words": ["Elon"]
+}

app.py

@@ -9,7 +9,7 @@ all_models = TTS.list_models()
 #     print(model)
 
 # print("Using model: ", all_models[0])
-model_name = all_models[0]
+model = all_models[0]
 # Init TTS
 
 
@@ -17,6 +17,8 @@ print("Downloading model...", '')
 
 voiceCloneModel = TTS('tts_models/multilingual/multi-dataset/your_tts')
 
+# tts = TTS("tts_models/de/thorsten/tacotron2-DDC")
+
 
 def run_cmd(command):
     try:
@@ -36,27 +38,35 @@ def inference(text, speaker):
         speaker = 'input/obama.mp3'
     elif (speaker == 'Speaker-4'):
         speaker = 'input/trump.wav'
+    elif (speaker == 'Rock'):
+        speaker = 'input/sample/Dwayne-Johnson-sample.wav'
+    elif (speaker == 'Elon'):
+        speaker = 'input/sample/Elon-Sample.wav'
+    elif (speaker == 'Obama'):
+        speaker = 'input/sample/Obama-sample.wav'
     else:
         speaker = 'input/z-default.wav'
     # print("speaker: ", speaker)
     # cmd = ['tts', '--text', text, '--out_path', 'output/tts_output.wav']
     # run_cmd(cmd)
     # Text to speech to a file
-    # tts = TTS(model_name="tts_models/multilingual/multi-dataset/your_tts",
+    # tts = TTS(model="tts_models/multilingual/multi-dataset/your_tts",
     #           progress_bar=False, gpu=True)
     voiceCloneModel.tts_to_file(text, speaker_wav=speaker,
                                 language="en", file_path="output/output.wav")
 
-    # for i in range(len(tts.languages)):
-    #     tts.tts_to_file(text=text,
-    #                     speaker=tts.speakers[i], language=tts.languages[0], file_path='output/output-'+str(i)+'.wav')
+    # for i in range(len(model.languages)):
+    #     model.tts_to_file(text=text,
+    #                       speaker=model.speakers[i], language=model.languages[0], file_path='output/output-'+str(i)+'.wav')
 
     return 'output/output.wav'
 
 
+sampleInput = ['Speaker-1', 'Speaker-2', 'Speaker-3',
+               'Speaker-4', 'Rock', 'Elon', 'Obama']
 inputs = [gr.inputs.Textbox(lines=5, label="Input Text"),
-          gr.inputs.Dropdown(['Speaker-1', 'Speaker-2', 'Speaker-3',
-                              'Speaker-4'], label="Model")
+          gr.inputs.Dropdown(
+              sampleInput, label="Sample Input", default="Speaker-1",)
           ]
 outputs = gr.outputs.Audio(type="filepath", label="Output Audio")
 title = "Text To Speech"

input/sample/Dwayne-Johnson-sample.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3c3d9d61a817cc459f890867c584a0496245c81b747205b48c18b49b6723d8b
+size 1578042

input/sample/Elon-Sample.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6609e4b4c5a8ce77eff174fd2eca989ded8a28fd2ca5b1866035fd54247649b
+size 1233978

input/sample/Obama-sample.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6dec4bb97019439e01a8909d68f5704e0e2c6de12c57865f3d4a01132519c045
+size 930874

output/output.wav

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ddd247a0b637133326079fa149c85c8544671c92c0c7a78ec93952db3cc0dfb2
-size 132172
+oid sha256:07abadaef514caf57a002e76da940e5900a6851ccba39076c3e7eb4359fdf82d
+size 1364620