Spaces:
Sleeping
Sleeping
File size: 1,968 Bytes
374b25a |
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 |
import streamlit as st
from TTS.api import TTS
import tempfile
import os
# Initialize TTS model (only once)
@st.cache_resource
def load_tts_model():
return TTS("tts_models/multilingual/multi-dataset/xtts_v2", gpu=True)
tts = load_tts_model()
# App title
st.title("🔊 Voice Cloning with XTTS v2")
# Text input
text_input = st.text_area("Enter the text you want to synthesize", height=150)
# Speaker file uploader
speaker_file = st.file_uploader("Upload a speaker WAV file", type=["wav"])
# Button to generate
if st.button("Generate Speech"):
if not text_input:
st.error("Please enter text.")
elif not speaker_file:
st.error("Please upload a speaker WAV file.")
else:
try:
with st.spinner("Generating voice..."):
# Save uploaded speaker audio temporarily
with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as speaker_temp:
speaker_temp.write(speaker_file.read())
speaker_path = speaker_temp.name
# Temporary file to store output
output_path = os.path.join(tempfile.gettempdir(), "output.wav")
# Generate speech
tts.tts_to_file(
text=text_input,
file_path=output_path,
speaker_wav=speaker_path,
language="en"
)
# Playback
st.audio(output_path, format="audio/wav")
# Download link
with open(output_path, "rb") as f:
st.download_button(
label="Download Audio",
data=f,
file_name="cloned_voice.wav",
mime="audio/wav"
)
# Clean up
os.remove(speaker_path)
except Exception as e:
st.error(f"An error occurred: {e}")
|