|
import gradio as gr |
|
import os |
|
import tempfile |
|
import re |
|
from pydub import AudioSegment |
|
from openai import OpenAI |
|
|
|
|
|
DEFAULT_API_KEY = "sk-GINVEtfNbrXNcGQhf3rEUIgzoicNGIApovqZxe0AYJF5PkTV" |
|
DEFAULT_BASE_URL = "https://open.keyai.shop" |
|
|
|
|
|
MAX_CHAR_LIMIT = 140964096 |
|
|
|
def clean_text(text): |
|
|
|
cleaned_text = re.sub(r'\s+', ' ', text.strip()) |
|
return cleaned_text |
|
|
|
def split_text(text, limit=MAX_CHAR_LIMIT): |
|
|
|
words = text.split(' ') |
|
chunks = [] |
|
current_chunk = "" |
|
for word in words: |
|
if len(current_chunk) + len(word) + 1 <= limit: |
|
current_chunk += word + " " |
|
else: |
|
chunks.append(current_chunk.strip()) |
|
current_chunk = word + " " |
|
if current_chunk: |
|
chunks.append(current_chunk.strip()) |
|
return chunks |
|
|
|
def tts(text, model, voice, speed): |
|
cleaned_text = clean_text(text) |
|
chunks = split_text(cleaned_text) |
|
|
|
audio_segments = [] |
|
|
|
try: |
|
client = OpenAI(api_key=DEFAULT_API_KEY, base_url=DEFAULT_BASE_URL + '/v1') |
|
for chunk in chunks: |
|
response = client.audio.speech.create( |
|
model=model, |
|
voice=voice, |
|
input=chunk, |
|
speed=speed |
|
) |
|
with tempfile.NamedTemporaryFile(suffix=".mp3", delete=False) as temp_file: |
|
temp_file.write(response.content) |
|
temp_file_path = temp_file.name |
|
audio_segments.append(AudioSegment.from_mp3(temp_file_path)) |
|
except Exception as error: |
|
raise gr.Error("Đã xảy ra lỗi khi tạo giọng nói. Vui lòng kiểm tra API key hoặc thử lại sau.") |
|
|
|
|
|
final_audio = sum(audio_segments) |
|
with tempfile.NamedTemporaryFile(suffix=".mp3", delete=False) as final_temp_file: |
|
final_audio.export(final_temp_file.name, format="mp3") |
|
final_audio_path = final_temp_file.name |
|
|
|
return final_audio_path |
|
|
|
|
|
custom_css = """ |
|
@import url('https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.0.0-beta3/css/all.min.css'); |
|
|
|
body { |
|
background-color: #1e1e2f; |
|
color: #e0e0e0; |
|
font-family: 'Arial', sans-serif; |
|
} |
|
.title { |
|
font-size: 2.5rem; |
|
font-weight: bold; |
|
text-align: center; |
|
margin-bottom: 30px; |
|
color: #f1f1f1; |
|
} |
|
.container { |
|
margin: auto; |
|
width: 90%; |
|
padding: 20px; |
|
} |
|
.card { |
|
background: #2e2e42; |
|
padding: 20px; |
|
border-radius: 8px; |
|
box-shadow: 0 4px 6px rgba(0, 0, 0, 0.5); |
|
margin-bottom: 20px; |
|
} |
|
.gradio-container { |
|
background: #1e1e2f; |
|
} |
|
button { |
|
background-color: #4CAF50; |
|
color: white; |
|
border: none; |
|
border-radius: 8px; |
|
padding: 10px 20px; |
|
font-size: 1rem; |
|
cursor: pointer; |
|
} |
|
button:hover { |
|
background-color: #45a049; |
|
} |
|
input, textarea, select { |
|
background: #3a3a50; |
|
border: 1px solid #555; |
|
color: #e0e0e0; |
|
padding: 8px; |
|
border-radius: 4px; |
|
} |
|
select { |
|
padding: 8px 10px; |
|
} |
|
label { |
|
font-weight: bold; |
|
margin-bottom: 5px; |
|
} |
|
""" |
|
|
|
with gr.Blocks(css=custom_css) as demo: |
|
|
|
gr.Markdown("<div class='title'><i class='fas fa-microphone-alt'></i> OpenAI TTS</div>") |
|
|
|
with gr.Column(elem_classes="container"): |
|
with gr.Row(): |
|
|
|
with gr.Column(elem_classes="card"): |
|
text = gr.Textbox(label="Văn bản đầu vào", placeholder="Nhập văn bản cần chuyển giọng nói...", lines=15) |
|
char_counter = gr.Markdown("Character count: 0") |
|
|
|
with gr.Column(elem_classes="card"): |
|
model = gr.Dropdown(choices=['tts-1', 'tts-1-hd'], label='Chọn Model', value='tts-1') |
|
voice = gr.Dropdown(choices=['alloy', 'echo', 'fable', 'onyx', 'nova', 'shimmer'], label='Chọn Giọng', value='alloy') |
|
speed = gr.Slider(minimum=0.5, maximum=2.0, step=0.1, label="Tốc độ", value=1.0) |
|
|
|
|
|
with gr.Row(elem_classes="card"): |
|
btn = gr.Button(" Chuyển ") |
|
|
|
|
|
with gr.Row(elem_classes="card"): |
|
output_audio = gr.Audio(label="Kết quả âm thanh") |
|
|
|
|
|
def update_char_counter(text): |
|
cleaned_text = clean_text(text) |
|
return f"Character count: {len(cleaned_text)}" |
|
|
|
text.change(fn=update_char_counter, inputs=text, outputs=char_counter) |
|
text.submit(fn=tts, inputs=[text, model, voice, speed], outputs=output_audio, api_name="tts_enter_key", concurrency_limit=None) |
|
btn.click(fn=tts, inputs=[text, model, voice, speed], outputs=output_audio, api_name="tts_button", concurrency_limit=None) |
|
|
|
demo.launch() |
|
|