Added new sample voices audio

.vscode/settings.json

@@ -1,3 +1,7 @@
 {
-  "cSpell.words": ["Elon"]
+  "cSpell.words": ["Elon"],
+  "[python]": {
+    "editor.defaultFormatter": "ms-python.autopep8"
+  },
+  "python.formatting.provider": "none"
 }

app.py

@@ -2,16 +2,19 @@ from subprocess import call
 import gradio as gr
 import os
 from TTS.api import TTS
+from TTS.utils.synthesizer import Synthesizer
 
 # List available 🐸TTS models and choose the first one
-all_models = TTS.list_models()
+# all_models = TTS.list_models()
 # for model in all_models:
 #     print(model)
 
 # print("Using model: ", all_models[0])
-model = all_models[0]
+# model = all_models[0]
 # Init TTS
 
+MAX_TXT_LEN = 500
+
 
 print("Downloading model...", '')
 
@@ -30,6 +33,10 @@ def run_cmd(command):
 
 
 def inference(text, speaker):
+    if len(text) > MAX_TXT_LEN:
+        text = text[:MAX_TXT_LEN]
+        print(
+            f"Input text was cutoff since it went over the {MAX_TXT_LEN} character limit.")
     if (speaker == 'Speaker-1'):
         speaker = 'input/amitabh.mp3'
     elif (speaker == 'Speaker-2'):
@@ -44,6 +51,10 @@ def inference(text, speaker):
         speaker = 'input/sample/Elon-Sample.wav'
     elif (speaker == 'Obama'):
         speaker = 'input/sample/Obama-sample.wav'
+    elif (speaker == 'Tony'):
+        speaker = 'input/sample/tony-stark-2.wav'
+    elif (speaker == 'Madara Uchiha'):
+        speaker = 'input/sample/Madara.mp3'
     else:
         speaker = 'input/z-default.wav'
     # print("speaker: ", speaker)
@@ -63,12 +74,12 @@ def inference(text, speaker):
 
 
 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-4', 'Rock', 'Elon', 'Obama', 'Tony', 'Madara Uchiha']
+inputs = [gr.Textbox(lines=5, label="Input Text"),
+          gr.Dropdown(
               sampleInput, label="Sample Input", default="Speaker-1",)
           ]
-outputs = gr.outputs.Audio(type="filepath", label="Output Audio")
+outputs = gr.Audio(type="filepath", label="Output Audio")
 title = "Text To Speech"
 description = "An example of using TTS to generate speech from text."
 article = ""

input/sample/Madara.mp3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:601c62ffb87f3879576194637190758defa9a13000c3fb38f814332a64e6835e
+size 259062

input/sample/tony-stark-2.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:106c43d952d7caf2d9e372cf4f74c7febde7a8a543532b5fc5e3a02ab4956cc3
+size 99862

input/sample/tony-stark.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a46f91b72ca5825e4166652af58155bb6c8e764b368b51da0652775a104c19fb
+size 17998

output/output.wav

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:07abadaef514caf57a002e76da940e5900a6851ccba39076c3e7eb4359fdf82d
-size 1364620
+oid sha256:d6cbd93714af794b46b7ccd96d30949ae39b55343bb841abf37a2d17b4ce4824
+size 172108