added project files

.dockerignore

@@ -0,0 +1,8 @@
+.gitignore
+.devcontainer
+__pycache__/
+*.git
+temp/
+archive/
+Pipfile
+Pipfile.lock

.gitignore

@@ -0,0 +1,166 @@
+archive/*
+temp/
+api_config.yml
+
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+data/*
+Python-3.11.6/
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#pdm.lock
+#   pdm stores project-wide configurations in .pdm.toml, but it is recommended to not include it
+#   in version control.
+#   https://pdm.fming.dev/#use-with-ide
+.pdm.toml
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+#.idea/

Dockerfile

@@ -0,0 +1,23 @@
+# Use an official Python runtime as a parent image
+FROM python:3.11.7-slim-bullseye
+
+RUN useradd -m -u 1000 user
+USER user
+ENV PATH="/home/user/.local/bin:$PATH"
+
+# Set the working directory in the container to /app
+WORKDIR /app
+
+# Copy the current directory contents into the container at /app
+COPY --chown=user . /app
+
+#Install ImageMagick
+RUN apt-get update && apt-get install -y imagemagick && sed -i '91d' /etc/ImageMagick-6/policy.xml
+# Install any needed packages specified in requirements.txt
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Make port 8000 available to the world outside this container
+EXPOSE 8000
+
+# Run main.py when the container launches
+CMD ["python", "main.py"]

README.md

@@ -8,4 +8,10 @@ pinned: false
 license: cc-by-nc-4.0
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+[//]: <> (Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference)
+
+## Multilang ASR Captioner
+
+A multilingual automatic speech recognition and video captioning tool using faster whisper. 
+
+Supports real-time translation to english. Runs on consumer grade cpu.

docker-compose.yml

@@ -0,0 +1,9 @@
+services:
+  app:
+    build: 
+      context: .
+      dockerfile: Dockerfile
+    ports:
+      - "8000:8000"
+    volumes:
+      - .:/app

main.py

@@ -0,0 +1,121 @@
+import shutil, os, logging, uvicorn, tempfile
+from typing import Optional
+
+from utils.process_video import process_video
+from utils.zip_response import zip_response
+from utils.api_configs import api_configs
+from utils.read_html import read_html
+from utils.logger import setup_logger
+
+from fastapi import FastAPI, UploadFile, HTTPException, Form, Depends
+from fastapi.responses import HTMLResponse, Response
+from fastapi.security import HTTPBasic
+from pydantic import BaseModel, field_validator
+
+app = FastAPI()
+security = HTTPBasic()
+api_configs_file = os.path.abspath("api_config.yml")
+
+class MP4Video(BaseModel):
+    video_file: UploadFile
+    
+    @property
+    def filename(self):
+        return self.video_file.filename
+    @property
+    def file(self):
+        return self.video_file.file
+
+    @field_validator('video_file')
+    def validate_video_file(cls, v):
+        if not v.filename.endswith('.mp4'):
+            raise HTTPException(status_code=500, detail='Invalid video file type. Please upload an MP4 file.')
+        return v
+
+class SRTFile(BaseModel):
+    srt_file: Optional[UploadFile] = None
+    
+    @property
+    def filename(self):
+        return self.srt_file.filename
+    @property
+    def file(self):
+        return self.srt_file.file
+    @property
+    def size(self):
+        return self.srt_file.size
+
+    @field_validator('srt_file')
+    def validate_srt_file(cls, v):
+        if v.size > 0 and not v.filename.endswith('.srt'):
+            raise HTTPException(status_code=422, detail='Invalid subtitle file type. Please upload an SRT file.')
+        return v
+
+@app.get("/")
+async def root():
+    html_content = f"""
+    {read_html(os.path.join(os.getcwd(),"static/landing_page.html"))}
+    """
+    return HTMLResponse(content=html_content)
+
+
+@app.get("/submit_video/")
+async def get_form():
+    html_content = f"""
+    {read_html(os.path.join(os.getcwd(),"static/submit_video.html"))}
+    """
+    return HTMLResponse(content=html_content)
+
+async def get_temp_dir():
+    dir = tempfile.TemporaryDirectory()
+    try:
+        yield dir.name
+    finally:
+        del dir
+
+@app.post("/process_video/")
+async def process_video_api(video_file: MP4Video = Depends(),
+                            srt_file: SRTFile = Depends(),
+                            task: Optional[str] = Form("transcribe"),
+                            max_words_per_line: Optional[int] = Form(6),
+                            fontsize: Optional[int] = Form(42),
+                            font: Optional[str] = Form("FuturaPTHeavy"),
+                            bg_color: Optional[str] = Form("#070a13b3"),
+                            text_color: Optional[str] = Form("white"),
+                            caption_mode: Optional[str] = Form("desktop"),
+                            temp_dir: str = Depends(get_temp_dir)
+                            ):
+    try:
+        logging.info("Creating temporary directories")
+        with open(os.path.join(temp_dir, video_file.filename), 'w+b') as temp_file:
+            logging.info("Copying video UploadFile to the temporary directory")
+            try:
+                shutil.copyfileobj(video_file.file, temp_file)
+            finally:
+                video_file.file.close()
+            logging.info("Copying SRT UploadFile to the temp_input_path")
+            if srt_file.size > 0:
+                with open(os.path.join(temp_dir, f"{video_file.filename.split('.')[0]}.srt"), 'w+b') as temp_srt_file:
+                    try:
+                        shutil.copyfileobj(srt_file.file, temp_srt_file)
+                    finally:
+                        srt_file.file.close()
+                logging.info("Processing the video...")
+                output_path, _ = process_video(temp_file.name, temp_srt_file.name, task, max_words_per_line, fontsize, font, bg_color, text_color, caption_mode)
+                logging.info("Zipping response...")
+                with open(os.path.join(temp_dir, f"{video_file.filename.split('.')[0]}.zip"), 'w+b') as temp_zip_file:
+                    zip_file = zip_response(temp_zip_file.name, [output_path, srt_path])
+                return Response(content = zip_file)
+            with open(os.path.join(temp_dir, f"{video_file.filename.split('.')[0]}.srt"), 'w+b') as temp_srt_file:
+                logging.info("Processing the video...")
+                output_path, srt_path = process_video(temp_file.name, None, task, max_words_per_line, fontsize, font, bg_color, text_color, caption_mode, api_configs_file)
+                logging.info("Zipping response...")
+                with open(os.path.join(temp_dir, f"{video_file.filename.split('.')[0]}.zip"), 'w+b') as temp_zip_file:
+                    zip_file = zip_response(temp_zip_file.name, [output_path, srt_path])
+                return Response(content = zip_file)  
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+    
+if __name__ == "__main__":
+    app_logger = setup_logger('appLogger', 'main.log', level=logging.DEBUG)
+    uvicorn.run(app, host="0.0.0.0", port=8000)

static/landing_page.html

@@ -0,0 +1,157 @@
+<!DOCTYPE html>
+<html>
+    <head>
+        <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.0.0-beta3/css/all.min.css" integrity="sha512-Fo3rlrZj/k7ujTnHg4CGR2D7kSs0v4LLanw2qksYuRlEzO+tcaEPQogQ0KaoGN26/zrn20ImR1DfuLWnOo7aBA==" crossorigin="anonymous" referrerpolicy="no-referrer" />
+        <style>
+            /* CSS Styles */
+            body {
+                font-family: 'Arial', sans-serif;
+                background-color: #f0f0f0;
+                color: #333;
+                line-height: 1.6;
+                text-align: center;
+                padding-top: 50px;
+                margin: 0;
+                height: 100vh;
+                display: flex;
+                align-items: center;
+                justify-content: center;
+                position: relative;
+            }
+
+            .container {
+                width: 90%;
+                max-width: 1000px; /* Increased max width */
+                margin: auto;
+                padding: 40px; /* Increased padding for more space */
+                background: #ffffff;
+                border-radius: 8px;
+                box-shadow: 0 5px 15px rgba(0, 0, 0, 0.15); /* Enhanced shadowing */
+                transform: translateY(-35%);
+            }
+
+            h1 {
+                color: #333;
+                margin-bottom: 20px;
+            }
+
+            p {
+                font-size: 18px;
+                color: #666;
+                margin-bottom: 30px;
+            }
+
+            .button {
+                display: inline-block;
+                padding: 15px 30px; /* Increased padding for larger buttons */
+                margin: 10px;
+                border-radius: 4px; /* Slightly more rounded corners */
+                color: white;
+                background-color: #4CAF50;
+                text-decoration: none;
+                font-size: 18px;
+                transition: background-color 0.3s, box-shadow 0.3s;
+            }
+
+            .button:hover {
+                background-color: #45a049;
+                box-shadow: 0 2px 10px rgba(0, 0, 0, 0.2); /* Subtle hover effect */
+            }
+
+            .button.docs {
+                background-color: #008CBA;
+            }
+
+            .button.docs:hover {
+                background-color: #007BAA;
+            }
+
+            .footer {
+                width: 100%;
+                background-color: #f0f0f0;
+                padding: 20px 0;
+                position: absolute;
+                bottom: 5rem;
+                text-align: center;
+            }
+
+            .footer a {
+                padding: 0.5rem;
+                text-decoration: none;
+            }
+            .fa-github:hover {
+                transform: scale(1.2)
+            }
+            .fa-github:hover {
+                transform: scale(1.2)
+            }
+            .fa-github{
+                color: #000000
+            }
+            .fa-linkedin:hover {
+                transform: scale(1.2)
+            }
+            .fa-linkedin {
+                color: #0077B5
+            }
+
+            /* Responsiveness */
+            @media (max-width: 768px) {
+                .container {
+                    width: 95%;
+                    padding: 20px;
+                    display: flex; /* Added to create a flex container */
+                    flex-direction: column; /* Stack elements vertically */
+                    align-items: center; /* Center-align items for a neat look */
+                }
+
+                h1 {
+                    font-size: 24px;
+                }
+
+                p {
+                    font-size: 16px;
+                    text-align: center; /* Center text for a balanced appearance */
+                }
+
+                .button {
+                    width: 80%; /* Set a specific width for both buttons */
+                    padding: 10px 20px;
+                    font-size: 16px;
+                    margin-bottom: 10px; /* Add some space between the buttons */
+                }
+
+                /* Ensure buttons are the same size */
+                .button.submit, .button.docs {
+                    width: calc(80% - 20px); /* Adjusting width to account for padding */
+                }
+            }
+
+            @media (max-height: 500px) {
+                body {
+                    padding-top: 20px;
+                    height: auto;
+                }
+
+                .container {
+                    align-items: center; /* Ensure center alignment in constrained height */
+                }
+            }
+        </style>
+    </head>
+    <body>
+        <div class="container">
+            <h1>Multilang-ASR-Captioner</h1>
+            <p>A multilingual automatic speech recognition and video captioning tool using faster whisper.</p>
+            <p>Supports real-time translation to english. Runs on consumer grade cpu.</p>
+            <a href="/submit_video" class="button submit">Submit Video</a>
+            <a href="/docs" class="button docs">Documentation</a>
+        </div>
+        <!-- Footer -->
+        <div class="footer">
+            <p>Created by:</p>
+            <a href="https://github.com/marquesafonso" class="github"><i class="fab fa-github fa-4x"></i></a>
+            <a href="https://www.linkedin.com/in/marquesafonso" class="linkedin"><i class="fab fa-linkedin fa-4x"></i></a>
+        </div>
+    </body>
+</html>

static/submit_video.html

@@ -0,0 +1,196 @@
+<!DOCTYPE html>
+<html>
+    <head>
+        <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.0.0-beta3/css/all.min.css" integrity="sha512-Fo3rlrZj/k7ujTnHg4CGR2D7kSs0v4LLanw2qksYuRlEzO+tcaEPQogQ0KaoGN26/zrn20ImR1DfuLWnOo7aBA==" crossorigin="anonymous" referrerpolicy="no-referrer" />
+        <style>
+            /* CSS Styles */
+            body {
+                font-family: 'Arial', sans-serif;
+                background-color: #f0f0f0;
+                color: #333;
+                line-height: 1.6;
+                margin: 0;
+                padding: 0;
+                display: flex;
+                flex-direction: column;
+                min-height: 100vh;
+            }
+
+            form {
+                max-width: 900px;
+                margin: .9rem auto;
+                padding: 1rem;
+                background: #ffffff;
+                border-radius: 8px;
+                box-shadow: 0 0 15px rgba(0, 0, 0, 0.1);
+                display: flex;
+                flex-direction: column;
+            }
+
+            .form-wrapper {
+                display: flex;
+                flex-wrap: wrap;
+                gap: 20px;
+            }
+
+            .form-group {
+                flex: 1;
+                min-width: calc(50% - 20px);
+                box-sizing: border-box;
+            }
+
+            .form-group h3 {
+                margin-bottom: 15px;
+                color: #4CAF50;
+                font-size: 18px;
+                border-bottom: 2px solid #4CAF50;
+                padding-bottom: 5px;
+            }
+
+            input[type=file],
+            input[type=number],
+            input[type=text],
+            select {
+                width: 100%;
+                padding: 10px;
+                margin-bottom: 10px;
+                border-radius: 4px;
+                border: 1px solid #ddd;
+                box-shadow: inset 0 1px 3px rgba(0, 0, 0, 0.1);
+                font-size: 13px;
+                box-sizing: border-box; /* Ensure padding and border are included in the element's total width and height */
+            }
+
+            input[type=submit] {
+                width: 100%;
+                background-color: #4CAF50;
+                color: white;
+                padding: 12px 18px;
+                border: none;
+                border-radius: 5px;
+                cursor: pointer;
+                font-size: 15px;
+                margin-top: 20px;
+                transition: background-color 0.3s ease;
+                box-sizing: border-box;
+            }
+
+            input[type=submit]:hover {
+                background-color: #45a049;
+            }
+
+            label {
+                margin-top: 10px;
+                display: block;
+                font-weight: bold;
+                font-size: 13px;
+            }
+
+            .footer {
+                width: 100%;
+                background-color: #f0f0f0;
+                text-align: center;
+            }
+
+            .footer a {
+                padding: 0.5rem;
+                text-decoration: none;
+            }
+
+            .fa-github:hover {
+                transform: scale(1.2);
+            }
+
+            .fa-github {
+                color: #000000;
+            }
+
+            .fa-linkedin:hover {
+                transform: scale(1.2);
+            }
+
+            .fa-linkedin {
+                color: #0077B5;
+            }
+
+            /* Additional Responsiveness */
+            @media (max-width: 992px) {
+                form {
+                    max-width: 90%;
+                    margin-left: 15%;
+                    margin-right: 15%; 
+                    padding: 15px;
+                }
+
+                .form-wrapper {
+                    flex-direction: column;
+                }
+
+                .form-group {
+                    min-width: 100%;
+                }
+            }
+
+            @media (max-width: 768px) {
+                form {
+                    max-width: 90%;
+                    margin-left: 10%;
+                    margin-right: 10%;
+                    padding: 15px;
+                }
+            }
+
+            @media (max-width: 480px) {
+                form {
+                    max-width: 90%;
+                    margin-left: 5%;
+                    margin-right: 5%;
+                    padding: 10px;
+                }
+            }
+        </style>
+    </head>
+    <body>
+        <form action="/process_video/" enctype="multipart/form-data" method="post">
+            <div class="form-wrapper">
+                <div class="form-group">
+                    <h3>Inputs & Task Selection</h3>
+                    <label for="video_file">Video File</label>
+                    <input type="file" id="video_file" name="video_file"><br>
+                    <label for="srt_file">Subtitles File</label>
+                    <input type="file" id="srt_file" name="srt_file"><br>
+                    <label for="task">Task</label>
+                    <select id="task" name="task">
+                        <option value="transcribe">Transcribe</option>
+                        <option value="translate">Translate</option>
+                    </select>
+                </div>
+                <div class="form-group">
+                    <h3>Visual Parameters</h3>
+                    <label for="max_words_per_line">Max words per line</label>
+                    <input type="number" id="max_words_per_line" name="max_words_per_line" value="6"><br>
+                    <label for="fontsize">Font size</label>
+                    <input type="number" id="fontsize" name="fontsize" value="42"><br>
+                    <label for="font">Font:</label>
+                    <input type="text" id="font" name="font" value="FuturaPTHeavy"><br>
+                    <label for="bg_color">Background color</label>
+                    <input type="text" id="bg_color" name="bg_color" value="#00FFFF00"><br>
+                    <label for="text_color">Text color</label>
+                    <input type="text" id="text_color" name="text_color" value="white"><br>
+                    <label for="caption_mode">Caption mode</label>
+                    <select id="caption_mode" name="caption_mode">
+                        <option value="desktop">Desktop</option>
+                        <option value="mobile">Mobile</option>
+                    </select>
+                </div>
+            </div>
+            <input type="submit" value="Submit">
+       </form>
+        <!-- Footer -->
+        <div class="footer">
+            <p>Created by:</p>
+            <a href="https://github.com/marquesafonso" class="github"><i class="fab fa-github fa-3x"></i></a>
+            <a href="https://www.linkedin.com/in/marquesafonso" class="linkedin"><i class="fab fa-linkedin fa-3x"></i></a>
+        </div>
+    </body>
+</html>

utils/api_configs.py

@@ -0,0 +1,6 @@
+import yaml
+
+def api_configs(config_file):
+    with open(config_file, 'r') as f:
+        db_config = yaml.safe_load(f)
+    return db_config["api_config"]

utils/logger.py

@@ -0,0 +1,22 @@
+import logging
+
+# https://signoz.io/guides/python-logging-best-practices/
+# Create and configure a named logger
+def setup_logger(name, log_file, level=logging.INFO):
+    logger = logging.getLogger(name)
+    logger.setLevel(level)
+
+    # Create handlers
+    file_handler = logging.FileHandler(log_file)
+    console_handler = logging.StreamHandler()
+
+    # Create formatters and add them to handlers
+    formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s')
+    file_handler.setFormatter(formatter)
+    console_handler.setFormatter(formatter)
+
+    # Add handlers to the logger
+    logger.addHandler(file_handler)
+    logger.addHandler(console_handler)
+
+    return logger

utils/process_video.py

@@ -0,0 +1,27 @@
+import logging, os
+from utils.transcriber import transcriber
+from utils.subtitler import subtitler
+
+def process_video(invideo_file: str,
+                  srt_file: str | None,
+                  task: str,
+                  max_words_per_line:int,
+                  fontsize:str,
+                  font:str,
+                  bg_color:str,
+                  text_color:str,
+                  caption_mode:str,
+                  config_file:str
+                  ):
+    invideo_path_parts = os.path.normpath(invideo_file).split(os.path.sep)
+    VIDEO_NAME = os.path.basename(invideo_file)
+    OUTVIDEO_PATH = os.path.join(os.path.normpath('/'.join(invideo_path_parts[:-1])), f"result_{VIDEO_NAME}")
+    if srt_file:
+        logging.info("Subtitling...")
+        subtitler(invideo_file, srt_file, OUTVIDEO_PATH, fontsize, font, bg_color, text_color, caption_mode)
+    else:
+        srt_file = os.path.normpath(f"{invideo_file.split('.')[0]}.srt")
+        transcriber(invideo_file, srt_file, max_words_per_line, task, config_file)
+        logging.info("Subtitling...")
+        subtitler(invideo_file, srt_file, OUTVIDEO_PATH, fontsize, font, bg_color, text_color, caption_mode)
+    return OUTVIDEO_PATH, srt_file

utils/read_html.py

@@ -0,0 +1,4 @@
+def read_html(html_file):
+    with open(html_file, 'r', encoding='utf-8') as f:
+        content = f.read()
+    return content

utils/subtitler.py

@@ -0,0 +1,59 @@
+from moviepy.editor import VideoFileClip, CompositeVideoClip, TextClip
+import os
+
+def parse_srt(srt_file):
+    """Parse the SRT file and return a list of (start, end, text) for each subtitle."""
+    with open(srt_file, "r", encoding='utf-8') as file:
+        lines = file.readlines()
+    i = 0
+    subtitles = []
+    while i < len(lines):
+        if lines[i].strip().isdigit():
+            timing_str = lines[i+1].strip().split(" --> ")
+            start = timing_str[0]
+            end = timing_str[1]
+            text = lines[i+2].strip()
+            subtitles.append((start, end, text))
+            i += 4
+        else:
+            i += 1
+    return subtitles
+
+def filter_caption_width(caption_mode:str):
+    if caption_mode == 'desktop':
+        caption_width_ratio = 0.5
+        caption_height_ratio = 0.8
+    elif caption_mode == 'mobile':
+        caption_width_ratio = 0.2
+        caption_height_ratio = 0.7
+    return caption_width_ratio, caption_height_ratio
+
+def subtitler(video_file:str,
+            srt_path:str,
+            output_file:str,
+            fontsize:int,
+            font: str,
+            bg_color:str,
+            text_color:str,
+            caption_mode:str
+            ):
+    """Add subtitles from an SRT file to a video."""
+    video_file = os.path.abspath(video_file)
+    srt_path = os.path.abspath(srt_path)
+    output_file = os.path.abspath(output_file)
+    clip = VideoFileClip(filename=video_file, target_resolution=None)
+    subtitles = parse_srt(srt_path)
+    subtitle_clips = []
+    caption_width_ratio, caption_height_ratio = filter_caption_width(caption_mode)
+    for start, end, text in subtitles:
+        # Create TextClip with specified styling
+        # To get a list of possible color and font values run: print(TextClip.list("font"), '\n\n', TextClip.list("color"))
+        txt_clip = TextClip(text, fontsize=fontsize, color=text_color, font=font, method='caption',
+                            bg_color=bg_color, align='center', size=(clip.w*caption_width_ratio, None))
+        txt_clip = txt_clip.set_position(('center', 'bottom')).set_duration(clip.duration).set_start(start).set_end(end)
+        subtitle_x_position = 'center'
+        subtitle_y_position = clip.h * caption_height_ratio
+        text_position = (subtitle_x_position, subtitle_y_position)                    
+        subtitle_clips.append(txt_clip.set_position(text_position))
+    video = CompositeVideoClip(size=None, clips=[clip] + subtitle_clips)
+    video.write_videofile(output_file, codec='libx264', audio_codec='aac')

utils/transcriber.py

@@ -0,0 +1,19 @@
+from gradio_client import Client, handle_file
+from utils.api_configs import api_configs
+import tempfile
+
+def transcriber(invideo_file:str, srt_file:str,
+        max_words_per_line:int, task:str,
+        config_file:str):
+        HF_TOKEN = api_configs(config_file)["secrets"]["hf-token"]
+        HF_SPACE = api_configs(config_file)["secrets"]["hf-space"]
+        client = Client(HF_SPACE, hf_token=HF_TOKEN)
+        result = client.predict(
+                video_input=handle_file(invideo_file),
+                max_words_per_line=max_words_per_line,
+                task=task,
+                api_name="/predict"
+        )
+        with open(srt_file, "w", encoding='utf-8') as file:
+                file.write(result[0])
+        return srt_file

utils/zip_response.py

@@ -0,0 +1,9 @@
+import zipfile, os
+
+def zip_response(temp_zip_file: str, files: list):
+    with zipfile.ZipFile(temp_zip_file, 'w') as zipf:
+        for file in files:
+            zipf.write(file, arcname=os.path.basename(file))
+    with open(temp_zip_file, 'rb') as zip_file:
+        zip_bytes = zip_file.read()
+    return zip_bytes