File size: 4,803 Bytes
8dc09d5 b6ace04 8dc09d5 b6ace04 8dc09d5 c9a73ce 8dc09d5 203a3ef 2bce58b 8dc09d5 47f799a b6ace04 8dc09d5 c9a73ce 8dc09d5 c9a73ce 8dc09d5 271f91e 8dc09d5 b6ace04 8dc09d5 c9a73ce 8dc09d5 |
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 |
import streamlit as st
import subprocess
import tempfile
import sys
import os
from os.path import exists
import requests
from huggingface_hub import HfFolder
import tarfile
from PIL import Image
HF_TOKEN = os.environ.get("HF_TOKEN")
# 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"
# TTS model files
URL_TTS_ONNX = "https://huggingface.co/CavidanZ/TTS-azerbaijani-model/resolve/main/last.onnx"
URL_TTS_ONNX_2 = "https://huggingface.co/CavidanZ/TTS-azerbaijani-model/resolve/main/last2.onnx"
TMP_PIPER_FILENAME = os.path.join(BASE_PATH, "piper.tgz")
headers = {"Authorization": f"Bearer {HF_TOKEN}"}
##########################
# 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: TTS model files #
#####################################################
# Create "Model" path if not existing
if os.path.exists(BASE_PATH_MODEL) == False:
os.makedirs(BASE_PATH_MODEL)
# --- Download TTS model --- #
response = requests.get(URL_TTS_ONNX, headers=headers)
if response.status_code == 200:
with open(os.path.join(BASE_PATH_MODEL, "last.onnx"), 'wb') as f:
f.write(response.content)
else:
st.markdown(f"Failed to download TTS from {URL_TTS_ONNX} (Status code: {response.status_code})")
response = requests.get((URL_TTS_ONNX + ".json"), headers=headers)
if response.status_code == 200:
with open(os.path.join(BASE_PATH_MODEL, "last.onnx.json"), 'wb') as f:
f.write(response.content)
else:
st.markdown(f"Failed to download TTS json from {URL_TTS_ONNX}.json (Status code: {response.status_code})")
response = requests.get(URL_TTS_ONNX_2, headers=headers)
if response.status_code == 200:
with open(os.path.join(BASE_PATH_MODEL, "last2.onnx"), 'wb') as f:
f.write(response.content)
else:
st.markdown(f"Failed to download TTS from {URL_TTS_ONNX} (Status code: {response.status_code})")
response = requests.get((URL_TTS_ONNX_2 + ".json"), headers=headers)
if response.status_code == 200:
with open(os.path.join(BASE_PATH_MODEL, "last2.onnx.json"), 'wb') as f:
f.write(response.content)
else:
st.markdown(f"Failed to download TTS json from {URL_TTS_ONNX}.json (Status code: {response.status_code})")
###########################
# 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('Salam!')
# st.header('Azərbaycan dilində mətni səsə çevirən model (TTS)')
# st.markdown('.')
st.title('Hello!')
st.header('A text-to-speech (TTS) model in Azerbaijani language')
with st.form("my_form"):
option = st.selectbox(
'Choose the model.',
['Model 1'],
['Model 2'])
text = st.text_area("Text to generate audio from:",max_chars=500)
submitted = st.form_submit_button("Submit")
if submitted:
with st.spinner("Please, wait... :)"):
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 "Model 1":
MODEL = "last.onnx"
case "Model 2":
MODEL = "last2.onnx"
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('Download audio', audio_bytes, file_name='TTS-Voice.wav')
except:
pass |