utils.py

import whisper
from transformers import MarianMTModel, MarianTokenizer, AutoTokenizer, AutoModelForSeq2SeqLM
import os
import tempfile
import subprocess

# Load Whisper model
model = whisper.load_model("base")

def process_video(video_file, language):
    # Save uploaded video to a temporary file
    temp_dir = tempfile.gettempdir()
    video_path = os.path.join(temp_dir, "input_video")  # No extension
    output_video_path = os.path.join(temp_dir, "converted_video.mp4")  # Convert to MP4 for compatibility

    try:
        # Save the uploaded file
        print("Saving uploaded video...")
        with open(video_path, "wb") as f:
            f.write(video_file.read())
        print(f"Video saved to {video_path}")

        # Convert the video to MP4 using ffmpeg
        print("Converting video to MP4...")
        subprocess.run(["ffmpeg", "-i", video_path, "-c:v", "libx264", "-preset", "fast", output_video_path], check=True)
        print(f"Video converted and saved to {output_video_path}")

        # Transcribe the video
        print("Transcribing video to English...")
        result = model.transcribe(output_video_path, language="en")
        print("Transcription completed!")

        # Translation logic
        segments = []
        if language == "English":
            segments = result["segments"]
        else:
            # Define translation models
            model_map = {
                "Hindi": "Helsinki-NLP/opus-mt-en-hi",
                "Spanish": "Helsinki-NLP/opus-mt-en-es",
                "French": "Helsinki-NLP/opus-mt-en-fr",
                "German": "Helsinki-NLP/opus-mt-en-de",
                "Telugu": "facebook/nllb-200-distilled-600M",
                "Portuguese": "Helsinki-NLP/opus-mt-en-pt",
                "Russian": "Helsinki-NLP/opus-mt-en-ru",
                "Chinese": "Helsinki-NLP/opus-mt-en-zh",
                "Arabic": "Helsinki-NLP/opus-mt-en-ar",
                "Japanese": "Helsinki-NLP/opus-mt-en-jap"
            }
            model_name = model_map.get(language)
            if not model_name:
                return f"Unsupported language: {language}"

            print(f"Loading translation model for {language}: {model_name}")
            if language == "Telugu":
                tokenizer = AutoTokenizer.from_pretrained(model_name)
                translation_model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
                tgt_lang = "tel_Telu"
                print(f"Translating to Telugu using NLLB-200 Distilled...")
                for segment in result["segments"]:
                    inputs = tokenizer(segment["text"], return_tensors="pt", padding=True)
                    translated_tokens = translation_model.generate(
                        **inputs, forced_bos_token_id=tokenizer.convert_tokens_to_ids(tgt_lang)
                    )
                    translated_text = tokenizer.batch_decode(translated_tokens, skip_special_tokens=True)[0]
                    segments.append({"text": translated_text, "start": segment["start"], "end": segment["end"]})
            else:
                tokenizer = MarianTokenizer.from_pretrained(model_name)
                translation_model = MarianMTModel.from_pretrained(model_name)
                print(f"Translating to {language}...")
                for segment in result["segments"]:
                    inputs = tokenizer(segment["text"], return_tensors="pt", padding=True)
                    translated = translation_model.generate(**inputs)
                    translated_text = tokenizer.decode(translated[0], skip_special_tokens=True)
                    segments.append({"text": translated_text, "start": segment["start"], "end": segment["end"]})

        # Create SRT file
        srt_path = os.path.join(tempfile.gettempdir(), "subtitles.srt")
        print(f"Creating SRT file at {srt_path}")
        with open(srt_path, "w", encoding="utf-8") as f:
            for i, segment in enumerate(segments, 1):
                start = f"{segment['start']:.3f}".replace(".", ",")
                end = f"{segment['end']:.3f}".replace(".", ",")
                text = segment["text"].strip()
                f.write(f"{i}\n00:00:{start} --> 00:00:{end}\n{text}\n\n")
        print("SRT file created successfully!")
        return srt_path

    except subprocess.CalledProcessError as e:
        return f"FFmpeg Error: {str(e)}"
    except Exception as e:
        return f"Unexpected Error: {str(e)}"
    finally:
        # Clean up temporary files
        print("Cleaning up temporary files...")
        if os.path.exists(video_path):
            os.remove(video_path)
        if os.path.exists(output_video_path):
            os.remove(output_video_path)