File size: 2,207 Bytes
3f53728
a13c70d
11fdcf5
 
 
 
 
 
 
 
a13c70d
3f53728
656d097
 
11fdcf5
a13c70d
 
656d097
 
 
 
 
d8bfda7
09f90bb
 
 
5d8ec27
09f90bb
5d8ec27
09f90bb
824e11d
656d097
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
import spaces
import gradio as gr
import torch
from TTS.api import TTS
import os
os.environ["COQUI_TOS_AGREED"] = "1"

device = "cuda"

tts = TTS("tts_models/multilingual/multi-dataset/xtts_v2").to(device)

@spaces.GPU(enable_queue=True)
def clone(text, audio, language):
    tts.tts_to_file(text=text, speaker_wav=audio, language=language, file_path="./output.wav")
    return "./output.wav"

iface = gr.Interface(fn=clone, 
                     inputs=[
                         gr.Textbox(label='Text'),
                         gr.Audio(type='filepath', label='Voice reference audio file'),
                         gr.Dropdown(label="Language", choices=["en", "es", "pt-br"], value="en")
                     ], 
                     outputs=gr.Audio(type='filepath'),
                     title='Voice Clone',
                     description="""
                     by [Tony Assi](https://www.tonyassi.com/)
                     This space uses xtts_v2 model. Non-commercial use only. [Coqui Public Model License](https://coqui.ai/cpml)
                     
                     Please ❤️ this Space. <a href="mailto: tony.assi.media@gmail.com">Email me</a>.
                     """,
                     theme = gr.themes.Base(primary_hue="teal",secondary_hue="teal",neutral_hue="slate"),
                     examples=[["Hey! It's me Dorthy, from the Wizard of Oz. Type in whatever you'd like me to say.","./audio/Wizard-of-Oz-Dorthy.wav", "en"],
                               ["It's me Vito Corleone, from the Godfather. Type in whatever you'd like me to say.","./audio/Godfather.wav", "en"],
                               ["Hey, it's me Paris Hilton. Type in whatever you'd like me to say.","./audio/Paris-Hilton.mp3", "en"],
                               ["Hey, it's me Megan Fox from Transformers. Type in whatever you'd like me to say.","./audio/Megan-Fox.mp3", "en"],
                               ["Hey there, it's me Jeff Goldblum. Type in whatever you'd like me to say.","./audio/Jeff-Goldblum.mp3", "en"],
                               ["Hey there, it's me Heath Ledger as the Joker. Type in whatever you'd like me to say.","./audio/Heath-Ledger.mp3", "en"],])
iface.launch()