File size: 3,034 Bytes
cf92f5c 4e243b0 cf92f5c 4e243b0 cf92f5c 4e243b0 cf92f5c 4e243b0 cf92f5c 4e243b0 cf92f5c 4e243b0 cf92f5c 4e243b0 2a7ff02 4e243b0 2a7ff02 4e243b0 cf92f5c 4e243b0 cf92f5c 4e243b0 cf92f5c |
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 |
import json
import torch
from TTS.api import TTS
import gradio as gr
from bs4 import BeautifulSoup
# Function to parse HTML and create a language dictionary
def parse_html_to_dict(file_path):
with open(file_path, 'r', encoding='utf-8') as file:
html_content = file.read()
soup = BeautifulSoup(html_content, 'html.parser')
language_dict = {}
for p in soup.find_all('p')[1:]:
parts = p.get_text().split()
if len(parts) >= 2:
code = parts[0].strip()
name = " ".join(parts[1:]).strip()
language_dict[name] = code
return language_dict
# Load the language dictionary from the HTML file
file_path = '1100_lan.htm' # Update this with your actual file path
language_dict = parse_html_to_dict(file_path)
languages = list(language_dict.keys())
iso_codes = list(language_dict.values())
# Get device
device = "cuda" if torch.cuda.is_available() else "cpu"
# Function to generate audio
def generate_audio(text, language_code):
model_name = f"tts_models/{language_code}/fairseq/vits"
tts = TTS(model_name=model_name, progress_bar=False).to(device)
output_path = "output_file.wav"
tts.tts_to_file(text=text, file_path=output_path)
return output_path
# Function to filter languages based on search term
def filter_languages(search_term):
if not search_term:
return gr.update(choices=languages, value=None)
filtered = [lang for lang in languages if search_term.lower() in lang.lower()]
return gr.update(choices=filtered, value=None if not filtered else filtered[0])
# Function to map selected language to ISO code
def update_iso_code(selected_language):
return language_dict.get(selected_language, "Unknown")
# Gradio interface
def run_interface():
with gr.Blocks() as demo:
with gr.Row():
search_input = gr.Textbox(
label="Search Language",
placeholder="Type to search...",
show_label=True
)
with gr.Row():
language_dropdown = gr.Dropdown(
choices=languages,
label="Select Language",
show_label=True,
interactive=True
)
with gr.Row():
text_input = gr.Textbox(label="Enter Text", lines=4, placeholder="Type your text here...")
generate_button = gr.Button("Generate Audio")
output_audio = gr.Audio(label="Generated Audio", type="filepath")
# Update dropdown choices when search input changes
search_input.change(
fn=filter_languages,
inputs=[search_input],
outputs=language_dropdown
)
# Generate audio when button is clicked
generate_button.click(
fn=lambda text, lang: generate_audio(text, update_iso_code(lang)),
inputs=[text_input, language_dropdown],
outputs=[output_audio]
)
demo.launch()
# Run the Gradio interface
if __name__ == "__main__":
run_interface()
|