Spaces:
Sleeping
Sleeping
update app
#1
by
Ionut-Bostan
- opened
app.py
CHANGED
@@ -10,7 +10,6 @@ predefined_texts = [
|
|
10 |
emotion_mapping = {"amused": 0, "anger": 1,
|
11 |
"disgust": 2, "neutral": 3, "sleepiness": 4}
|
12 |
|
13 |
-
|
14 |
def synthesize_speech(input_type, text, own_text, speaker_id, embed_type, emotion_id):
|
15 |
if input_type == "Choose from examples":
|
16 |
selected_text = text
|
@@ -26,20 +25,21 @@ def synthesize_speech(input_type, text, own_text, speaker_id, embed_type, emotio
|
|
26 |
audio_file = f'output/result/EmoV_DB/{selected_text}.wav'
|
27 |
return audio_file
|
28 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
29 |
|
30 |
iface = gr.Interface(
|
31 |
fn=synthesize_speech,
|
32 |
-
inputs=[
|
33 |
-
|
34 |
-
["Choose from examples", "Enter your own text"], label="Input Type"),
|
35 |
-
gr.inputs.Dropdown(choices=predefined_texts, label="Select a text"),
|
36 |
-
gr.inputs.Textbox(lines=2, label="Enter your own text"),
|
37 |
-
gr.inputs.Slider(minimum=0, maximum=3, step=1,
|
38 |
-
default=0, label="Speaker ID"),
|
39 |
-
gr.inputs.Radio(["bert_embed", "emotion_id"], label="Embedding Type"),
|
40 |
-
gr.inputs.Dropdown(choices=emotion_mapping, label="Select Emotion"),
|
41 |
-
],
|
42 |
-
outputs=gr.outputs.Audio(type="filepath"),
|
43 |
title="Text-to-Speech Demo",
|
|
|
44 |
)
|
|
|
45 |
iface.launch()
|
|
|
|
10 |
emotion_mapping = {"amused": 0, "anger": 1,
|
11 |
"disgust": 2, "neutral": 3, "sleepiness": 4}
|
12 |
|
|
|
13 |
def synthesize_speech(input_type, text, own_text, speaker_id, embed_type, emotion_id):
|
14 |
if input_type == "Choose from examples":
|
15 |
selected_text = text
|
|
|
25 |
audio_file = f'output/result/EmoV_DB/{selected_text}.wav'
|
26 |
return audio_file
|
27 |
|
28 |
+
input_type = gr.Radio(
|
29 |
+
choices=["Choose from examples", "Enter your own text"], label="Input Type")
|
30 |
+
text = gr.Dropdown(choices=predefined_texts, label="Select a text")
|
31 |
+
own_text = gr.Textbox(lines=2, label="Enter your own text")
|
32 |
+
speaker_id = gr.Slider(minimum=0, maximum=3, step=1, default=0, label="Speaker ID")
|
33 |
+
embed_type = gr.Radio(choices=["bert_embed", "emotion_id"], label="Embedding Type")
|
34 |
+
emotion_id = gr.Dropdown(choices=list(emotion_mapping.keys()), label="Select Emotion")
|
35 |
|
36 |
iface = gr.Interface(
|
37 |
fn=synthesize_speech,
|
38 |
+
inputs=[input_type, text, own_text, speaker_id, embed_type, emotion_id],
|
39 |
+
outputs=gr.Audio(type="filepath"),
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
40 |
title="Text-to-Speech Demo",
|
41 |
+
description="Select or enter text and configure options to synthesize speech."
|
42 |
)
|
43 |
+
|
44 |
iface.launch()
|
45 |
+
|