|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
import logging |
|
import gradio as gr |
|
|
|
title = "# Next-gen Kaldi: Generate subtitles for videos" |
|
|
|
description = """ |
|
This space shows how to generate subtitles/captions with Next-gen Kaldi. |
|
|
|
It is running on CPU within a docker container provided by Hugging Face. |
|
|
|
See more information by visiting the following links: |
|
|
|
- <https://github.com/k2-fsa/sherpa-onnx> |
|
- <https://github.com/k2-fsa/icefall> |
|
- <https://github.com/k2-fsa/k2> |
|
- <https://github.com/lhotse-speech/lhotse> |
|
|
|
If you want to deploy it locally, please see |
|
<https://k2-fsa.github.io/sherpa/> |
|
""" |
|
|
|
|
|
|
|
css = """ |
|
.result {display:flex;flex-direction:column} |
|
.result_item {padding:15px;margin-bottom:8px;border-radius:15px;width:100%} |
|
.result_item_success {background-color:mediumaquamarine;color:white;align-self:start} |
|
.result_item_error {background-color:#ff7070;color:white;align-self:start} |
|
""" |
|
|
|
|
|
def build_html_output(s: str, style: str = "result_item_success"): |
|
return f""" |
|
<div class='result'> |
|
<div class='result_item {style}'> |
|
{s} |
|
</div> |
|
</div> |
|
""" |
|
|
|
|
|
def process_uploaded_file( |
|
language: str, |
|
in_filename: str, |
|
): |
|
if in_filename is None or in_filename == "": |
|
return "", build_html_output( |
|
"Please first upload a file and then click " |
|
'the button "submit for recognition"', |
|
"result_item_error", |
|
) |
|
|
|
logging.info(f"Processing uploaded file: {in_filename}") |
|
|
|
return "Done", build_html_output("ok", "result_item_success") |
|
|
|
|
|
demo = gr.Blocks(css=css) |
|
|
|
|
|
with demo: |
|
gr.Markdown(title) |
|
language_choices = ["English", "Chinese"] |
|
|
|
language_radio = gr.Radio( |
|
label="Language", |
|
choices=language_choices, |
|
value=language_choices[0], |
|
) |
|
|
|
with gr.Tabs(): |
|
with gr.TabItem("Upload video from disk"): |
|
uploaded_file = gr.Video( |
|
source="upload", |
|
interactive=True, |
|
type="filepath", |
|
label="Upload from disk", |
|
) |
|
upload_button = gr.Button("Submit for recognition") |
|
uploaded_output = gr.Textbox(label="Recognized speech from uploaded file") |
|
uploaded_html_info = gr.HTML(label="Info") |
|
|
|
upload_button.click( |
|
process_uploaded_file, |
|
inputs=[ |
|
language_radio, |
|
uploaded_file, |
|
], |
|
outputs=[uploaded_output, uploaded_html_info], |
|
) |
|
|
|
gr.Markdown(description) |
|
|
|
if __name__ == "__main__": |
|
formatter = "%(asctime)s %(levelname)s [%(filename)s:%(lineno)d] %(message)s" |
|
|
|
logging.basicConfig(format=formatter, level=logging.INFO) |
|
|
|
demo.launch() |
|
|
