4月30日学习进度

import uuid

import threading

import re

import os

import tempfile

import soundfile as sf

from flask import Blueprint, jsonify, send_file, current_app

from pathlib import Path

from pptx import Presentation

from pptx.util import Inches

from werkzeug.utils import secure_filename

from F5_TTS.src.f5_tts.infer.utils_infer import (

    preprocess_ref_audio_text,

    infer_process,

    remove_silence_for_generated_wav

)

video_bp = Blueprint('video_bp', __name__)

video_tasks = {}

#@video_bp.route('/upload', methods=['POST'])

def video_upload(ppt_file,script_file,F5TTS_ema_model,vocoder,ref_audio_path):

    task_id = str(uuid.uuid4())

    video_tasks[task_id] = {

        'status': 'uploading',

        'progress': 0,

        'message': 'Uploading files...',

        'files': {}

    }

    # 创建任务目录

    save_dir = Path(current_app.config['VIDEO_UPLOAD_FOLDER']) / task_id

    save_dir.mkdir(parents=True, exist_ok=True)

    # 处理PPT文件

    if ppt_file.filename != '':

        ppt_filename = secure_filename(ppt_file.filename)

        ppt_path = save_dir / ppt_filename

        ppt_file.save(ppt_path)

        video_tasks[task_id]['files']['ppt'] = str(ppt_path)

    # 处理讲解脚本

    if script_file.filename != '':

        script_filename = secure_filename(script_file.filename)

        script_path = save_dir / script_filename

        script_file.save(script_path)

        video_tasks[task_id]['files']['script'] = str(script_path)

    # 更新任务状态

    video_tasks[task_id].update({

        'status': 'processing',

        'progress': 20,

        'message': 'Processing files...'

    })

    #ref_audio_path = "data/voice/"+ request.form['ref_audio_name']

    # 开启后台线程处理任务

    app_obj = current_app._get_current_object()

    threading.Thread(target=process_video_task, args=(app_obj, task_id, F5TTS_ema_model,vocoder,ref_audio_path)).start()

    return task_id

def process_video_task(app, task_id,F5TTS_ema_model,vocoder,ref_audio_path):

    try:

        with app.app_context():

            task = video_tasks[task_id]

            update_video_task(task_id, 30, 'Validating files...')

            ppt_path = Path(task['files'].get('ppt', ''))

            script_path = Path(task['files'].get('script', ''))

            if not ppt_path.exists():

                raise FileNotFoundError("PPT file not found")

            if not script_path.exists():

                raise FileNotFoundError("Script file not found")

            update_video_task(task_id, 40, 'Parsing script...')

            page_contents = parse_script(script_path)

            update_video_task(task_id, 60, 'Generating audio...')

            audio_dir = ppt_path.parent / "audio"

            audio_files = generate_audio_files(page_contents, audio_dir,F5TTS_ema_model,vocoder,ref_audio_path)

            update_video_task(task_id, 80, 'Inserting audio...')

            output_path = ppt_path.parent / f"audio_{ppt_path.name}"

            insert_audio_to_ppt(ppt_path, audio_files, output_path)

            video_tasks[task_id].update({

                'status': 'completed',

                'progress': 100,

                'result_path': str(output_path),

                'message': 'Processing complete'

            })

    except Exception as e:

        video_tasks[task_id].update({

            'status': 'failed',

            'progress': 100,

            'message': f'Error: {str(e)}'

        })

def update_video_task(task_id, progress, message):

    video_tasks[task_id].update({

        'progress': progress,

        'message': message

    })

def parse_script(txt_path):

    with open(txt_path, "r", encoding="utf-8") as f:

        lines = f.readlines()

    page_contents = {}

    current_page = None

    page_pattern = re.compile(r"第\s*(\d+)\s*页")

    for line in lines:

        line = line.strip()

        match = page_pattern.match(line)

        if match:

            current_page = int(match.group(1))

            page_contents[current_page] = ""

        elif current_page is not None:

            page_contents[current_page] += line + " "

    return page_contents

def convert_to_chinese_date(text):

    """Convert dates and numbers in the text to Chinese format."""

    num_map = {"0": "零", "1": "一", "2": "二", "3": "三", "4": "四",

               "5": "五", "6": "六", "7": "七", "8": "八", "9": "九"}

   

    def number_to_chinese(match):

        number = match.group()

        if len(number) == 1:  # 单个数字

            return num_map[number]

        elif len(number) == 2:  # 两位数

            if number.startswith("1"):  # 特殊处理10-19

                return "十" + (num_map[number[1]] if number[1] != "0" else "")

            else:

                return num_map[number[0]] + "十" + (num_map[number[1]] if number[1] != "0" else "")

        else:

            return "".join(num_map[digit] for digit in number)  # 处理三位及以上的数字

    # 将日期格式（如12月、10日）处理为中文读法

    text = re.sub(r'\d+', number_to_chinese, text)

    return text

def generate_audio_files(page_contents, output_dir,F5TTS_ema_model,vocoder,ref_audio_path):

    output_dir.mkdir(exist_ok=True)

    audio_files = {}

    for page, text in page_contents.items():

        audio_path = output_dir / f"page_{page}.mp3"

        try:

            gen_text = convert_to_chinese_date(text)

            ref_audio_data, ref_text = preprocess_ref_audio_text(ref_audio_path, '')

            final_wave, final_sample_rate, _ = infer_process(

                ref_audio_data,

                ref_text,

                gen_text,

                F5TTS_ema_model,

                vocoder,

                cross_fade_duration=0.15,

                speed=1.0,

            )

            # Remove silences from generated audio

            with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as temp_generated_audio:

                sf.write(temp_generated_audio.name, final_wave, final_sample_rate)

                remove_silence_for_generated_wav(temp_generated_audio.name)

                final_wave, _ = sf.read(temp_generated_audio.name)

                sf.write(audio_path, final_wave, final_sample_rate)

           

            audio_files[page] = audio_path

        except Exception as e:

            raise RuntimeError(f"Audio generation failed (Page {page}): {str(e)}")

    return audio_files

def insert_audio_to_ppt(ppt_path, audio_files, output_path):

    prs = Presentation(ppt_path)

    for page, audio_path in audio_files.items():

        if page <= len(prs.slides):

            slide = prs.slides[page-1]

            # 添加音频图标

            left = Inches(1)

            top = Inches(3)

            width = Inches(1)

            height = Inches(1)

            audio_shape = slide.shapes.add_movie(

                str(audio_path),

                left, top, width, height,

                poster_frame_image=None,

                mime_type='audio/mp3'

            )

            # 设置自动播放

            try:

                movie_element = audio_shape._element.find(".//p:movie")

                if movie_element is not None:

                    movie_element.set("playMode", "auto")

            except Exception as e:

                current_app.logger.warning(f"Auto-play setup failed: {str(e)}")

    prs.save(output_path)

@video_bp.route('/status/<task_id>')

def video_status(task_id):

    return jsonify(video_tasks.get(task_id, {'status': 'not_found'}))

@video_bp.route('/download/<task_id>')

def video_download(task_id):

    task = video_tasks.get(task_id)

    if not task or task['status'] != 'completed':

        return jsonify({'error': 'File not available'}), 404

    file_path = os.path.join("D:/Project/VisualStudioCode/EduVoice/",str(task['result_path']).replace('\\','/'))

    return send_file(file_path,as_attachment=True)