Create app.py

app.py

@@ -0,0 +1,67 @@
+import gradio as gr
+
+title = "fairseq S^2: A Scalable and Integrable Speech Synthesis Toolkit"
+
+description = "Gradio Demo for fairseq S^2: A Scalable and Integrable Speech Synthesis Toolkit. To use it, simply add your text, or click one of the examples to load them. Read more at the links below."
+
+article = "<p style='text-align: center'><a href='https://arxiv.org/abs/2109.06912' target='_blank'>fairseq S^2: A Scalable and Integrable Speech Synthesis Toolkit</a> | <a href='https://github.com/pytorch/fairseq/tree/main/examples/speech_synthesis' target='_blank'>Github Repo</a></p>"
+
+examples = [
+    ["Hello this is a test run","fastspeech2-en-200_speaker-cv4"],
+    ["Hello, this is a test run.","tts_transformer-en-200_speaker-cv4"],
+    ["Bonjour, ceci est un test.","tts_transformer-fr-cv7_css10"],
+    ["BЗдравствуйте, это пробный запуск.","tts_transformer-ru-cv7_css10"],
+    ["Merhaba, bu bir deneme çalışmasıdır.","tts_transformer-tr-cv7"],
+    ["Xin chào, đây là một cuộc chạy thử nghiệm.","tts_transformer-vi-cv7"],
+    ["مرحبًا ، هذا اختبار تشغيل.","tts_transformer-ar-cv7"],
+    ["Hola, esta es una prueba.","tts_transformer-es-css10"]
+]
+
+io1 = gr.Interface.load("huggingface/facebook/fastspeech2-en-200_speaker-cv4")
+
+io2 = gr.Interface.load("huggingface/facebook/tts_transformer-en-200_speaker-cv4")
+
+io3 = gr.Interface.load("huggingface/facebook/tts_transformer-fr-cv7_css10")
+
+io4 = gr.Interface.load("huggingface/facebook/tts_transformer-ru-cv7_css10")
+
+io5 = gr.Interface.load("huggingface/facebook/tts_transformer-tr-cv7")
+
+io6 = gr.Interface.load("huggingface/facebook/tts_transformer-vi-cv7")
+
+io7 = gr.Interface.load("huggingface/facebook/tts_transformer-ar-cv7")
+
+io8 = gr.Interface.load("huggingface/facebook/tts_transformer-es-css10")
+
+
+    
+def inference(text,model):
+   if model == "fastspeech2-en-200_speaker-cv4":
+        outtext = io1(text)
+   elif model == "tts_transformer-en-200_speaker-cv4":
+        outtext = io2(text)
+   elif model == "tts_transformer-fr-cv7_css10":
+        outtext = io3(text)
+   elif model == "tts_transformer-ru-cv7_css10":
+        outtext = io4(text)
+   elif model == "tts_transformer-tr-cv7":
+        outtext = io5(text)
+   elif model == "tts_transformer-vi-cv7":
+        outtext = io6(text)
+   elif model == "tts_transformer-ar-cv7":
+        outtext = io7(text)
+   else:
+        outtext = io8(text)
+   return outtext 
+
+
+gr.Interface(
+    inference, 
+    [gr.inputs.Textbox(label="Input",lines=5),gr.inputs.Dropdown(choices=["fastspeech2-en-200_speaker-cv4","tts_transformer-en-200_speaker-cv4","tts_transformer-tr-cv7","tts_transformer-fr-cv7_css10","tts_transformer-ru-cv7_css10","tts_transformer-vi-cv7","tts_transformer-ar-cv7","tts_transformer-es-css10"], type="value", default="fastspeech2-en-200_speaker-cv4", label="model")
+],
+    gr.outputs.Audio(label="Output"),
+    examples=examples,
+    article=article,
+    title=title,
+    description=description,
+    enable_queue=True).launch()