File size: 4,306 Bytes
32b2aaa
 
 
 
627d3d7
32b2aaa
 
bf13828
32b2aaa
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f83b1b7
32b2aaa
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
bf13828
32b2aaa
 
 
 
f83b1b7
32b2aaa
 
 
 
 
 
 
 
 
 
 
 
bf13828
32b2aaa
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
import gradio as gr
import torch
from modules.speaker import Speaker
from modules.utils.SeedContext import SeedContext
from modules.utils.hf import spaces
from modules.models import load_chat_tts
from modules.utils.rng import np_rng
from modules.webui import webui_config
from modules.webui.webui_utils import get_speakers, tts_generate

import tempfile

names_list = [
    "Alice",
    "Bob",
    "Carol",
    "Carlos",
    "Charlie",
    "Chuck",
    "Chad",
    "Craig",
    "Dan",
    "Dave",
    "David",
    "Erin",
    "Eve",
    "Yves",
    "Faythe",
    "Frank",
    "Grace",
    "Heidi",
    "Ivan",
    "Judy",
    "Mallory",
    "Mallet",
    "Darth",
    "Michael",
    "Mike",
    "Niaj",
    "Olivia",
    "Oscar",
    "Peggy",
    "Pat",
    "Rupert",
    "Sybil",
    "Trent",
    "Ted",
    "Trudy",
    "Victor",
    "Vanna",
    "Walter",
    "Wendy",
]


@torch.inference_mode()
@spaces.GPU
def create_spk_from_seed(
    seed: int,
    name: str,
    gender: str,
    desc: str,
):
    chat_tts = load_chat_tts()
    with SeedContext(seed, True):
        emb = chat_tts.sample_random_speaker()
    spk = Speaker(seed=-2, name=name, gender=gender, describe=desc)
    spk.emb = emb

    with tempfile.NamedTemporaryFile(delete=False, suffix=".pt") as tmp_file:
        torch.save(spk, tmp_file)
        tmp_file_path = tmp_file.name

    return tmp_file_path


@torch.inference_mode()
@spaces.GPU
def test_spk_voice(seed: int, text: str):
    return tts_generate(
        spk=seed,
        text=text,
    )


def random_speaker():
    seed = np_rng()
    name = names_list[seed % len(names_list)]
    return seed, name


def speaker_creator_ui():
    def on_generate(seed, name, gender, desc):
        file_path = create_spk_from_seed(seed, name, gender, desc)
        return file_path

    def create_test_voice_card(seed_input):
        with gr.Group():
            gr.Markdown("🎤Test voice")
            with gr.Row():
                test_voice_btn = gr.Button("Test Voice", variant="secondary")

                with gr.Column(scale=4):
                    test_text = gr.Textbox(
                        label="Test Text",
                        placeholder="Please input test text",
                        value=webui_config.localization.DEFAULT_SPEAKER_TEST_TEXT,
                    )
                    with gr.Row():
                        current_seed = gr.Label(label="Current Seed", value=-1)
                        with gr.Column(scale=4):
                            output_audio = gr.Audio(label="Output Audio", format="mp3")

        test_voice_btn.click(
            fn=test_spk_voice,
            inputs=[seed_input, test_text],
            outputs=[output_audio],
        )
        test_voice_btn.click(
            fn=lambda x: x,
            inputs=[seed_input],
            outputs=[current_seed],
        )

    gr.Markdown("SPEAKER_CREATOR_GUIDE")

    with gr.Row():
        with gr.Column(scale=2):
            with gr.Group():
                gr.Markdown("ℹ️Speaker info")
                seed_input = gr.Number(label="Seed", value=2)
                name_input = gr.Textbox(
                    label="Name", placeholder="Enter speaker name", value="Bob"
                )
                gender_input = gr.Textbox(
                    label="Gender", placeholder="Enter gender", value="*"
                )
                desc_input = gr.Textbox(
                    label="Description",
                    placeholder="Enter description",
                )
                random_button = gr.Button("Random Speaker")
            with gr.Group():
                gr.Markdown("🔊Generate speaker.pt")
                generate_button = gr.Button("Save .pt file")
                output_file = gr.File(label="Save to File")
        with gr.Column(scale=5):
            create_test_voice_card(seed_input=seed_input)
            create_test_voice_card(seed_input=seed_input)
            create_test_voice_card(seed_input=seed_input)
            create_test_voice_card(seed_input=seed_input)

    random_button.click(
        random_speaker,
        outputs=[seed_input, name_input],
    )

    generate_button.click(
        fn=on_generate,
        inputs=[seed_input, name_input, gender_input, desc_input],
        outputs=[output_file],
    )