Update app.py
Browse files
app.py
CHANGED
|
@@ -281,41 +281,40 @@ class FileManager:
|
|
| 281 |
except Exception:
|
| 282 |
pass # Ignore if directory isn't empty or can't be removed
|
| 283 |
|
| 284 |
-
|
| 285 |
-
|
| 286 |
if not file_path:
|
| 287 |
return None
|
| 288 |
-
|
| 289 |
filename = os.path.basename(file_path)
|
| 290 |
return f"""
|
| 291 |
-
<
|
| 292 |
-
|
| 293 |
-
|
| 294 |
-
|
| 295 |
-
|
| 296 |
-
|
| 297 |
-
|
| 298 |
-
|
| 299 |
-
|
| 300 |
-
|
| 301 |
-
|
| 302 |
-
|
| 303 |
-
|
| 304 |
-
|
| 305 |
-
|
| 306 |
-
|
| 307 |
-
|
| 308 |
-
|
| 309 |
-
|
| 310 |
-
|
| 311 |
-
|
| 312 |
-
</a>
|
| 313 |
-
</div>
|
| 314 |
"""
|
| 315 |
|
| 316 |
# Create global file manager
|
| 317 |
file_manager = FileManager()
|
| 318 |
|
|
|
|
|
|
|
| 319 |
# IMPROVEMENT 3: Parallel Processing for Segments
|
| 320 |
async def generate_accurate_srt(
|
| 321 |
text: str,
|
|
@@ -462,7 +461,7 @@ async def process_text_with_progress(
|
|
| 462 |
):
|
| 463 |
# Input validation
|
| 464 |
if not text or text.strip() == "":
|
| 465 |
-
return None, None, None, "Please enter some text to convert to speech."
|
| 466 |
|
| 467 |
# Format pitch and rate strings
|
| 468 |
pitch_str = f"{pitch:+d}Hz" if pitch != 0 else "+0Hz"
|
|
@@ -486,28 +485,14 @@ async def process_text_with_progress(
|
|
| 486 |
parallel=parallel_processing
|
| 487 |
)
|
| 488 |
|
| 489 |
-
#
|
| 490 |
-
|
| 491 |
-
audio_download_link = create_download_link(audio_path, "Download Audio (.mp3)")
|
| 492 |
-
|
| 493 |
-
# Return exactly 4 values in the expected order:
|
| 494 |
-
# 1. Audio preview path
|
| 495 |
-
# 2. SRT download button HTML
|
| 496 |
-
# 3. Audio download button HTML
|
| 497 |
-
# 4. Status message
|
| 498 |
-
return (
|
| 499 |
-
audio_path, # Audio preview
|
| 500 |
-
srt_download_link, # SRT download button
|
| 501 |
-
audio_download_link, # Audio download button
|
| 502 |
-
"Generation completed successfully!" # Status message
|
| 503 |
-
)
|
| 504 |
-
|
| 505 |
except TTSError as e:
|
| 506 |
-
# Return
|
| 507 |
-
return None, None, None, f"TTS Error: {str(e)}"
|
| 508 |
except Exception as e:
|
| 509 |
-
# Return
|
| 510 |
-
return None, None, None, f"Unexpected error: {str(e)}"
|
| 511 |
|
| 512 |
# Voice options dictionary
|
| 513 |
voice_options = {
|
|
@@ -557,41 +542,33 @@ import atexit
|
|
| 557 |
atexit.register(file_manager.cleanup_all)
|
| 558 |
|
| 559 |
# Create Gradio interface
|
| 560 |
-
with gr.Blocks(title="Advanced TTS with Configurable SRT Generation"
|
| 561 |
gr.Markdown("# Advanced TTS with Configurable SRT Generation")
|
| 562 |
gr.Markdown("Generate perfectly synchronized audio and subtitles with natural speech patterns.")
|
| 563 |
|
| 564 |
with gr.Row():
|
| 565 |
with gr.Column(scale=3):
|
| 566 |
-
text_input = gr.Textbox(
|
| 567 |
-
label="Enter Text",
|
| 568 |
-
lines=10,
|
| 569 |
-
placeholder="Enter your text here...",
|
| 570 |
-
elem_id="text_input"
|
| 571 |
-
)
|
| 572 |
|
| 573 |
with gr.Column(scale=2):
|
| 574 |
voice_dropdown = gr.Dropdown(
|
| 575 |
label="Select Voice",
|
| 576 |
choices=list(voice_options.keys()),
|
| 577 |
-
value="Jenny Female"
|
| 578 |
-
elem_id="voice_dropdown"
|
| 579 |
)
|
| 580 |
pitch_slider = gr.Slider(
|
| 581 |
label="Pitch Adjustment (Hz)",
|
| 582 |
minimum=-10,
|
| 583 |
maximum=10,
|
| 584 |
value=0,
|
| 585 |
-
step=1
|
| 586 |
-
elem_id="pitch_slider"
|
| 587 |
)
|
| 588 |
rate_slider = gr.Slider(
|
| 589 |
label="Rate Adjustment (%)",
|
| 590 |
minimum=-25,
|
| 591 |
maximum=25,
|
| 592 |
value=0,
|
| 593 |
-
step=1
|
| 594 |
-
elem_id="rate_slider"
|
| 595 |
)
|
| 596 |
|
| 597 |
with gr.Row():
|
|
@@ -602,8 +579,7 @@ with gr.Blocks(title="Advanced TTS with Configurable SRT Generation", analytics_
|
|
| 602 |
maximum=12,
|
| 603 |
value=6,
|
| 604 |
step=1,
|
| 605 |
-
info="Controls how many words appear on each line of the subtitle"
|
| 606 |
-
elem_id="words_per_line"
|
| 607 |
)
|
| 608 |
with gr.Column():
|
| 609 |
lines_per_segment = gr.Slider(
|
|
@@ -612,34 +588,30 @@ with gr.Blocks(title="Advanced TTS with Configurable SRT Generation", analytics_
|
|
| 612 |
maximum=4,
|
| 613 |
value=2,
|
| 614 |
step=1,
|
| 615 |
-
info="Controls how many lines appear in each subtitle segment"
|
| 616 |
-
elem_id="lines_per_segment"
|
| 617 |
)
|
| 618 |
with gr.Column():
|
| 619 |
parallel_processing = gr.Checkbox(
|
| 620 |
label="Enable Parallel Processing",
|
| 621 |
value=True,
|
| 622 |
-
info="Process multiple segments simultaneously for faster conversion (recommended for longer texts)"
|
| 623 |
-
elem_id="parallel_processing"
|
| 624 |
)
|
| 625 |
|
| 626 |
-
submit_btn = gr.Button("Generate Audio & Subtitles"
|
| 627 |
-
# Output components
|
| 628 |
-
with gr.Column(variant="panel") as output_col:
|
| 629 |
-
gr.Markdown("### Generated Output")
|
| 630 |
-
# Main output containers
|
| 631 |
-
with gr.Row():
|
| 632 |
-
audio_preview = gr.Audio(label="Preview Audio", elem_id="audio_preview")
|
| 633 |
-
|
| 634 |
-
with gr.Row():
|
| 635 |
-
with gr.Column(scale=1):
|
| 636 |
-
srt_download = gr.HTML(label="", elem_id="srt_download")
|
| 637 |
-
with gr.Column(scale=1):
|
| 638 |
-
audio_download = gr.HTML(label="", elem_id="audio_download")
|
| 639 |
-
|
| 640 |
-
error_output = gr.Textbox(label="Status", elem_id="error_output")
|
| 641 |
|
| 642 |
-
#
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 643 |
submit_btn.click(
|
| 644 |
fn=process_text_with_progress,
|
| 645 |
inputs=[
|
|
@@ -652,11 +624,15 @@ with gr.Blocks(title="Advanced TTS with Configurable SRT Generation", analytics_
|
|
| 652 |
parallel_processing
|
| 653 |
],
|
| 654 |
outputs=[
|
| 655 |
-
|
| 656 |
-
|
| 657 |
-
|
| 658 |
-
error_output
|
| 659 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 660 |
)
|
| 661 |
|
| 662 |
if __name__ == "__main__":
|
|
|
|
| 281 |
except Exception:
|
| 282 |
pass # Ignore if directory isn't empty or can't be removed
|
| 283 |
|
| 284 |
+
|
| 285 |
+
def create_download_link(file_path, label="Download"):
|
| 286 |
if not file_path:
|
| 287 |
return None
|
|
|
|
| 288 |
filename = os.path.basename(file_path)
|
| 289 |
return f"""
|
| 290 |
+
<a href="file={file_path}"
|
| 291 |
+
download="{filename}"
|
| 292 |
+
target="_blank"
|
| 293 |
+
rel="noopener noreferrer"
|
| 294 |
+
style="display: inline-block; padding: 10px 20px; background: linear-gradient(135deg, #4776E6, #8E54E9); color: white; text-decoration: none; border-radius: 8px; font-weight: 600; transition: all 0.3s ease; text-align: center; width: 100%; margin: 5px 0;"
|
| 295 |
+
onmouseover="this.style.transform='translateY(-2px)'; this.style.boxShadow='0 5px 15px rgba(71, 118, 230, 0.3)';"
|
| 296 |
+
onmouseout="this.style.transform='translateY(0)'; this.style.boxShadow='none';"
|
| 297 |
+
onclick="event.preventDefault(); fetch(this.href).then(resp => resp.blob()).then(blob => {{
|
| 298 |
+
const url = window.URL.createObjectURL(blob);
|
| 299 |
+
const a = document.createElement('a');
|
| 300 |
+
a.style.display = 'none';
|
| 301 |
+
a.href = url;
|
| 302 |
+
a.download = '{filename}';
|
| 303 |
+
document.body.appendChild(a);
|
| 304 |
+
a.click();
|
| 305 |
+
window.URL.revokeObjectURL(url);
|
| 306 |
+
document.body.removeChild(a);
|
| 307 |
+
}});">
|
| 308 |
+
⬇️ {label}
|
| 309 |
+
</a>
|
| 310 |
+
|
|
|
|
|
|
|
| 311 |
"""
|
| 312 |
|
| 313 |
# Create global file manager
|
| 314 |
file_manager = FileManager()
|
| 315 |
|
| 316 |
+
|
| 317 |
+
|
| 318 |
# IMPROVEMENT 3: Parallel Processing for Segments
|
| 319 |
async def generate_accurate_srt(
|
| 320 |
text: str,
|
|
|
|
| 461 |
):
|
| 462 |
# Input validation
|
| 463 |
if not text or text.strip() == "":
|
| 464 |
+
return None, None, None, True, "Please enter some text to convert to speech."
|
| 465 |
|
| 466 |
# Format pitch and rate strings
|
| 467 |
pitch_str = f"{pitch:+d}Hz" if pitch != 0 else "+0Hz"
|
|
|
|
| 485 |
parallel=parallel_processing
|
| 486 |
)
|
| 487 |
|
| 488 |
+
# If successful, return results and hide error
|
| 489 |
+
return srt_path, audio_path, audio_path, False, ""
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 490 |
except TTSError as e:
|
| 491 |
+
# Return specific TTS error
|
| 492 |
+
return None, None, None, True, f"TTS Error: {str(e)}"
|
| 493 |
except Exception as e:
|
| 494 |
+
# Return any other error
|
| 495 |
+
return None, None, None, True, f"Unexpected error: {str(e)}"
|
| 496 |
|
| 497 |
# Voice options dictionary
|
| 498 |
voice_options = {
|
|
|
|
| 542 |
atexit.register(file_manager.cleanup_all)
|
| 543 |
|
| 544 |
# Create Gradio interface
|
| 545 |
+
with gr.Blocks(title="Advanced TTS with Configurable SRT Generation") as app:
|
| 546 |
gr.Markdown("# Advanced TTS with Configurable SRT Generation")
|
| 547 |
gr.Markdown("Generate perfectly synchronized audio and subtitles with natural speech patterns.")
|
| 548 |
|
| 549 |
with gr.Row():
|
| 550 |
with gr.Column(scale=3):
|
| 551 |
+
text_input = gr.Textbox(label="Enter Text", lines=10, placeholder="Enter your text here...")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 552 |
|
| 553 |
with gr.Column(scale=2):
|
| 554 |
voice_dropdown = gr.Dropdown(
|
| 555 |
label="Select Voice",
|
| 556 |
choices=list(voice_options.keys()),
|
| 557 |
+
value="Jenny Female"
|
|
|
|
| 558 |
)
|
| 559 |
pitch_slider = gr.Slider(
|
| 560 |
label="Pitch Adjustment (Hz)",
|
| 561 |
minimum=-10,
|
| 562 |
maximum=10,
|
| 563 |
value=0,
|
| 564 |
+
step=1
|
|
|
|
| 565 |
)
|
| 566 |
rate_slider = gr.Slider(
|
| 567 |
label="Rate Adjustment (%)",
|
| 568 |
minimum=-25,
|
| 569 |
maximum=25,
|
| 570 |
value=0,
|
| 571 |
+
step=1
|
|
|
|
| 572 |
)
|
| 573 |
|
| 574 |
with gr.Row():
|
|
|
|
| 579 |
maximum=12,
|
| 580 |
value=6,
|
| 581 |
step=1,
|
| 582 |
+
info="Controls how many words appear on each line of the subtitle"
|
|
|
|
| 583 |
)
|
| 584 |
with gr.Column():
|
| 585 |
lines_per_segment = gr.Slider(
|
|
|
|
| 588 |
maximum=4,
|
| 589 |
value=2,
|
| 590 |
step=1,
|
| 591 |
+
info="Controls how many lines appear in each subtitle segment"
|
|
|
|
| 592 |
)
|
| 593 |
with gr.Column():
|
| 594 |
parallel_processing = gr.Checkbox(
|
| 595 |
label="Enable Parallel Processing",
|
| 596 |
value=True,
|
| 597 |
+
info="Process multiple segments simultaneously for faster conversion (recommended for longer texts)"
|
|
|
|
| 598 |
)
|
| 599 |
|
| 600 |
+
submit_btn = gr.Button("Generate Audio & Subtitles")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 601 |
|
| 602 |
+
# Add error message component
|
| 603 |
+
error_output = gr.Textbox(label="Status", visible=False)
|
| 604 |
+
|
| 605 |
+
with gr.Row():
|
| 606 |
+
with gr.Column():
|
| 607 |
+
audio_output = gr.Audio(label="Preview Audio")
|
| 608 |
+
with gr.Column():
|
| 609 |
+
srt_file = gr.File(label="Download SRT")
|
| 610 |
+
audio_file = gr.File(label="Download Audio")
|
| 611 |
+
srt_download = gr.HTML(elem_classes="download-btn")
|
| 612 |
+
audio_download = gr.HTML(elem_classes="download-btn")
|
| 613 |
+
|
| 614 |
+
# Handle button click with manual error handling instead of .catch()
|
| 615 |
submit_btn.click(
|
| 616 |
fn=process_text_with_progress,
|
| 617 |
inputs=[
|
|
|
|
| 624 |
parallel_processing
|
| 625 |
],
|
| 626 |
outputs=[
|
| 627 |
+
srt_file,
|
| 628 |
+
audio_file,
|
| 629 |
+
audio_output,
|
| 630 |
+
error_output,
|
| 631 |
+
error_output,
|
| 632 |
+
srt_download,
|
| 633 |
+
audio_download
|
| 634 |
+
],
|
| 635 |
+
api_name="generate"
|
| 636 |
)
|
| 637 |
|
| 638 |
if __name__ == "__main__":
|