salfter/whisper-webui
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

added NLLB translation

Browse Source
This commit is contained in:
jhj0517
2023-03-27 21:28:47 +09:00
parent 4e82fb64b3
commit 3fde2e0f02
4 changed files with 84 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

29
app.py
View File

@@ -1,7 +1,8 @@
import gradio as gr
from modules.whisper_Inference import WhisperInference
from modules.nllb_inference import NLLBInference
import os
from ui.htmls import CSS, MARKDOWN
from ui.htmls import *
from modules.youtube_manager import get_ytmetas
@@ -21,6 +22,7 @@ def on_change_models(model_size):
whisper_inf = WhisperInference()
nllb_inf = NLLBInference()
block = gr.Blocks(css=CSS).queue(api_open=False)
with block:
@@ -100,4 +102,29 @@ with block:
btn_openfolder.click(fn=lambda: open_fodler("outputs"), inputs=None, outputs=None)
dd_model.change(fn=on_change_models, inputs=[dd_model], outputs=[cb_translate])
with gr.TabItem("T2T Translation"): # tab 4
with gr.Row():
file_subs = gr.Files(type="file", label="Upload Subtitle Files to translate here",
file_types=['.vtt', '.srt'])
with gr.TabItem("NLLB"): # sub tab1
with gr.Row():
dd_nllb_model = gr.Dropdown(label="Model", value=nllb_inf.default_model_size,
choices=nllb_inf.available_models)
dd_nllb_sourcelang = gr.Dropdown(label="Source Language", choices=nllb_inf.available_source_langs)
dd_nllb_targetlang = gr.Dropdown(label="Target Language", choices=nllb_inf.available_target_langs)
with gr.Row():
btn_run = gr.Button("TRANSLATE SUBTITLE FILE", variant="primary")
with gr.Row():
tb_indicator = gr.Textbox(label="Output")
btn_openfolder = gr.Button('📂').style(full_width=False)
with gr.Column():
md_vram_table = gr.HTML(NLLB_VRAM_TABLE, elem_id="md_nllb_vram_table")
btn_run.click(fn=nllb_inf.translate_file,
inputs=[file_subs, dd_nllb_model, dd_nllb_sourcelang, dd_nllb_targetlang],
outputs=[tb_indicator])
btn_openfolder.click(fn=lambda: open_fodler("outputs\\translations"), inputs=None, outputs=None)
block.launch()

2
modules/nllb_inference.py
View File

@@ -68,7 +68,7 @@ class NLLBInference:
write_file(subtitle, f"{output_path}.srt")
elif file_path == ".vtt":
elif file_ext == ".vtt":
parsed_dicts = parse_vtt(file_path=file_path)
total_progress = len(parsed_dicts)
for index, dic in enumerate(parsed_dicts):

0
outputs/translations/outputs for translation are saved here.txt Normal file
View File

55
ui/htmls.py
View File

@@ -39,4 +39,59 @@ CSS = """
MARKDOWN = """
### [Whisper Web-UI](https://github.com/jhj0517/Whsiper-WebUI)
"""
NLLB_VRAM_TABLE = """
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<style>
table {
border-collapse: collapse;
width: 100%;
}
th, td {
border: 1px solid #dddddd;
text-align: left;
padding: 8px;
}
th {
background-color: #f2f2f2;
}
</style>
</head>
<body>
<details>
<summary>VRAM usage for each model</summary>
<table>
<thead>
<tr>
<th>Model name</th>
<th>Required VRAM</th>
</tr>
</thead>
<tbody>
<tr>
<td>nllb-200-3.3B</td>
<td>~16GB</td>
</tr>
<tr>
<td>nllb-200-1.3B</td>
<td>~8GB</td>
</tr>
<tr>
<td>nllb-200-distilled-600M</td>
<td>~4GB</td>
</tr>
</tbody>
</table>
<p><strong>Note:</strong> Be mindful of your VRAM! The table above provides an approximate VRAM usage for each model.</p>
</details>
</body>
</html>
"""