Hugging Face's logo

Spaces:
Thorsten-Voice
/
TTS
Running

App Files Community
TTS
File size: 7,041 Bytes 
8b53cc6
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4cad777
 
8b53cc6
4cad777
 
 
 
 
 
 
8b53cc6
 
 
 
 
 
 
 
 
 
 
 
4cad777
8b53cc6
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4cad777
 
 
 
 
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
164
165
166
167
168
169
 
import streamlit as st
import subprocess
import tempfile
import sys
import os
from os.path import exists
import requests
import tarfile
from PIL import Image

# Set base path
BASE_PATH = os.getcwd() # /home/user/app
BASE_PATH_MODEL = os.path.join(BASE_PATH, "Model")

# Piper TTS download url
URL_PIPER_DOWNLOAD = "https://github.com/rhasspy/piper/releases/download/v1.2.0/piper_amd64.tar.gz"

# Thorsten-Voice TTS model files
URL_TV_HOCHDEUTSCH_ONNX = "https://huggingface.co/rhasspy/piper-voices/resolve/v1.0.0/de/de_DE/thorsten/high/de_DE-thorsten-high.onnx"
URL_TV_EMOTIONAL_ONNX = "https://huggingface.co/rhasspy/piper-voices/resolve/v1.0.0/de/de_DE/thorsten_emotional/medium/de_DE-thorsten_emotional-medium.onnx"
URL_TV_HESSISCH_ONNX = "https://huggingface.co/Thorsten-Voice/Hessisch/resolve/main/Thorsten-Voice_Hessisch_Piper_high-Oct2023.onnx"

TMP_PIPER_FILENAME = os.path.join(BASE_PATH, "piper.tgz")

##########################
# CHECK OR INSTALL PIPER #
##########################
if os.path.exists(os.path.join(BASE_PATH,"piper")) == False:

    # Piper not downloaded and extracted yet, let's do this first.
    response = requests.get(URL_PIPER_DOWNLOAD)

    if response.status_code == 200:
        with open(TMP_PIPER_FILENAME, 'wb') as f:
            f.write(response.content)

        with tarfile.open(TMP_PIPER_FILENAME, 'r:gz') as tar:
            tar.extractall(BASE_PATH)

    else:
        st.markdown(f"Failed to download Piper TTS from {URL_PIPER_DOWNLOAD} (Status code: {response.status_code})")


#####################################################
# CHECK OR DOWNLOAD: All Thorsten-Voice model files #
#####################################################

# Create "Model" path if not existing
if os.path.exists(BASE_PATH_MODEL) == False:
    os.makedirs(BASE_PATH_MODEL)

    # --- Download "NEUTRAL" TTS model --- #
    response = requests.get(URL_TV_HOCHDEUTSCH_ONNX) 
    if response.status_code == 200:
        with open(os.path.join(BASE_PATH_MODEL, "TV-Neutral.onnx"), 'wb') as f:
            f.write(response.content)

    response = requests.get(URL_TV_HOCHDEUTSCH_ONNX + ".json") 
    if response.status_code == 200:
        with open(os.path.join(BASE_PATH_MODEL, "TV-Neutral.onnx.json"), 'wb') as f:
            f.write(response.content)


    # --- Download "EMOTIONAL" TTS model --- #
    response = requests.get(URL_TV_EMOTIONAL_ONNX) 
    if response.status_code == 200:
        with open(os.path.join(BASE_PATH_MODEL, "TV-Emotional.onnx"), 'wb') as f:
            f.write(response.content)

    response = requests.get(URL_TV_EMOTIONAL_ONNX + ".json") 
    if response.status_code == 200:
        with open(os.path.join(BASE_PATH_MODEL, "TV-Emotional.onnx.json"), 'wb') as f:
            f.write(response.content)


    # --- Download "HESSISCH" TTS model --- #
    response = requests.get(URL_TV_HESSISCH_ONNX) 
    if response.status_code == 200:
        with open(os.path.join(BASE_PATH_MODEL, "TV-Hessisch.onnx"), 'wb') as f:
            f.write(response.content)

    response = requests.get(URL_TV_HESSISCH_ONNX + ".json") 
    if response.status_code == 200:
        with open(os.path.join(BASE_PATH_MODEL, "TV-Hessisch.onnx.json"), 'wb') as f:
            f.write(response.content)


###########################
# MODEL DOWNLOAD FINISHED #
###########################

hide_streamlit_style = """
            <style>
            #MainMenu {visibility: hidden;}
            header {visibility: hidden;}
            footer {visibility: hidden;}
            .st-emotion-cache-1y4p8pa {padding-top: 0rem;}
            </style>
            """
st.markdown(hide_streamlit_style, unsafe_allow_html=True)

st.title('Guude! Thorsten-Voice hier 👋')
st.header('Deutsche, kostenlose und hochwertige KI Sprachausgabe (TTS)')

#st.subheader('Wofür?')
st.markdown('Meine künstlichen [Thorsten-Voice](https://www.Thorsten-Voice.de) Stimmen sind für **Content Creator** (Voice Over) geeignet (Youtube Videos/Shorts, Instagram, Tik Tok, ...).' +
            ' Für **Schulen**, **Wissenschaft/Forschung, kommerzielle Produkte**,' +
            ' **Hobby und Bastelprojekte** und natürlich auch gerne **einfach nur zum Spaß** um witzige **Sprachnachrichten** zu verschicken.')
st.markdown('Hier kannst Du meine künstliche Stimme in verschiedenen Emotionen und mit (Süd)Hessischem Dialekt direkt ausprobieren 😊. Das gefällt Dir? Dann kannst Du meine Stimme' +
            ' auch **lokal** und **kostenlos** auf deinem **eigenen Computer** (Windows, Linux, Mac OS und Raspberry Pi OS) benutzen.')
st.markdown('Mehr Infos auf der [Projektwebseite](https://www.Thorsten-Voice.de) und auf meinem [Youtube Kanal](https://www.youtube.com/c/thorstenmueller)')

#st.image('Thorsten-Voice_transparent_klein.png')

with st.form("my_form"):
   option = st.selectbox(
     'Wie soll die Stimme klingen?',
     ('Normal', 'Fröhlich', 'Wütend', 'Angewidert', 'Betrunken', 'Schläfrig', 'Flüsternd', 'Hessischer Dialekt'))
   
   text = st.text_area("Zu sprechender Text",max_chars=500)
   submitted = st.form_submit_button("Sprechen")
   
   if submitted:
    with st.spinner("Bitte einen Augenblick Geduld - bin gleich soweit ..."):
        filename = tempfile.NamedTemporaryFile(suffix=".wav", delete=False)

        # Set Piper TTS command based on choice
        PIPER_CMD = os.path.join(BASE_PATH,"piper","piper")
        SPEAKER_ID = "0"

        match option:
            case "Normal":
                MODEL = "TV-Neutral.onnx"
            case "Hessischer Dialekt":
                MODEL = "TV-Hessisch.onnx"
            case "Fröhlich":
                MODEL = "TV-Emotional.onnx"
                SPEAKER_ID = "0"
            case "Wütend":
                MODEL = "TV-Emotional.onnx"
                SPEAKER_ID = "1"
            case "Angewidert":
                MODEL = "TV-Emotional.onnx"
                SPEAKER_ID = "2"
            case "Betrunken":
                MODEL = "TV-Emotional.onnx"
                SPEAKER_ID = "3"
            case "Schläfrig":
                MODEL = "TV-Emotional.onnx"
                SPEAKER_ID = "5"
            case "Flüsternd":
                MODEL = "TV-Emotional.onnx"
                SPEAKER_ID = "7"

        cmd = "echo '" + text + "' | " + BASE_PATH + "/piper/piper --model " + os.path.join(BASE_PATH_MODEL, MODEL) + " --speaker " + SPEAKER_ID + " --output_file " + filename.name

        result = subprocess.run(cmd, shell=True)
        audio_file = open(filename.name, 'rb')
        audio_bytes = audio_file.read()
        st.audio(audio_bytes,format="audio/wav")
try:
    st.download_button('Runterladen', audio_bytes, file_name='Thorsten-Voice.wav')
except:
    pass

#st.subheader('Mehr Infos?')
#st.markdown('Möchtest Du mehr über die Möglichkeiten von **Thorsten-Voice** erfahren? Beispielsweise wie Du meine Stimme' +
#            ' auch komplett ohne Internet auf deinem PC mit Windows, Linux, Mac OS oder auf dem Raspberry Pi verwendest.'+ 
#            ' Dann sieh gerne auf der Projektwebseite www.Thorsten-Voice.de nach.')