EasyDub 配音視頻生成平臺：SpringBoot + Thymeleaf + Spring AI 實戰(zhàn)開發(fā)

作者：編程疏影 2025-05-15 08:21:16

通過整合 Spring Boot、Thymeleaf、Redis、FFmpeg 與 AI 模型接口（Whisper、XTTSv2 等），我們構(gòu)建了一個功能強大且易用的 EasyDub Web 配音系統(tǒng)，支持異步處理、狀態(tài)輪詢、數(shù)字人合成與完整視頻輸出。?

本項目旨在構(gòu)建一個 Web 端一鍵生成 AI 配音視頻的系統(tǒng)，提供從“上傳視頻 → 提取語音 → 翻譯 → 合成音頻 → 合成字幕與數(shù)字人 → 下載結(jié)果”的完整流程。后端基于 SpringBoot，前端使用 Thymeleaf + Bootstrap，結(jié)合 Redis 實現(xiàn)異步任務(wù)狀態(tài)跟蹤與進(jìn)度輪詢，支持多用戶并發(fā)任務(wù)處理。

功能亮點

?? 全流程：上傳原視頻 → 翻譯 → 配音合成 → 視頻輸出
?? Spring AI：調(diào)用 AI 模型實現(xiàn)翻譯、合成
??? Web UI：Thymeleaf + Bootstrap 實現(xiàn)進(jìn)度輪詢
?? Redis + Spring Task 實現(xiàn)異步任務(wù)與進(jìn)度管理
?? 實際 DEMO：上傳 original_video.mp4 → 下載 linly_dubbing.mp4

項目結(jié)構(gòu)

com.icoderoad.easydub
├── controller
│   └── DubbingController.java
├── service
│   ├── DubbingService.java
│   └── ProgressService.java
├── config
│   └── TaskConfig.java
├── model
│   └── TaskStatus.java
├── templates
│   └── index.html
├── static
│   └── bootstrap + js
├── application.yml
└── EasyDubApplication.java

SpringBoot 構(gòu)建 REST 接口

視頻上傳與任務(wù)創(chuàng)建接口

package com.icoderoad.easydub.controller;


import com.icoderoad.easydub.service.DubbingService;
import com.icoderoad.easydub.service.ProgressService;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.web.bind.annotation.*;
import org.springframework.web.multipart.MultipartFile;


@RestController
@RequestMapping("/api")
public class DubbingController {


    @Autowired
    private DubbingService dubbingService;


    @PostMapping("/upload")
    public String upload(@RequestParam("file") MultipartFile file) {
        return dubbingService.handleUpload(file);
    }


    @GetMapping("/progress/{taskId}")
    public String getProgress(@PathVariable String taskId) {
        return dubbingService.getProgress(taskId);
    }


    @GetMapping("/download/{taskId}")
    public String getDownloadUrl(@PathVariable String taskId) {
        return dubbingService.getDownloadUrl(taskId);
    }
}

Spring Task + Redis 實現(xiàn)任務(wù)調(diào)度

配置異步線程池

package com.icoderoad.easydub.config;


import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;
import org.springframework.scheduling.concurrent.ThreadPoolTaskExecutor;


import java.util.concurrent.Executor;


@Configuration
public class TaskConfig {
    @Bean(name = "taskExecutor")
    public Executor taskExecutor() {
        ThreadPoolTaskExecutor executor = new ThreadPoolTaskExecutor();
        executor.setCorePoolSize(3);
        executor.setMaxPoolSize(10);
        executor.setQueueCapacity(100);
        executor.setThreadNamePrefix("DubbingTask-");
        executor.initialize();
        return executor;
    }
}

后臺任務(wù)處理服務(wù)

package com.icoderoad.easydub.service;


import com.icoderoad.easydub.model.TaskStatus;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.core.io.ClassPathResource;
import org.springframework.scheduling.annotation.Async;
import org.springframework.stereotype.Service;
import org.springframework.web.multipart.MultipartFile;
import redis.clients.jedis.Jedis;


import java.io.File;
import java.io.FileOutputStream;
import java.util.UUID;


@Service
public class DubbingService {


    @Autowired
    private ProgressService progressService;


    private final String baseDir = "output/";


    public String handleUpload(MultipartFile file) {
        String taskId = UUID.randomUUID().toString();
        File saveFile = new File(baseDir + taskId + "_original.mp4");
        try (FileOutputStream fos = new FileOutputStream(saveFile)) {
            fos.write(file.getBytes());
        } catch (Exception e) {
            return "上傳失敗：" + e.getMessage();
        }


        progressService.init(taskId);
        processAsync(taskId, saveFile.getAbsolutePath());
        return taskId;
    }


    @Async("taskExecutor")
    public void processAsync(String taskId, String inputPath) {
        try {
            progressService.update(taskId, "提取音頻中...");
            String audioPath = extractAudio(inputPath);


            progressService.update(taskId, "識別翻譯中...");
            String translatedText = callSpringAIWhisperAndTranslate(audioPath);


            progressService.update(taskId, "合成語音中...");
            String newVoice = synthesizeAudio(translatedText);


            progressService.update(taskId, "合成視頻中...");
            String finalVideo = composeVideo(inputPath, newVoice, taskId);


            progressService.complete(taskId, finalVideo);
        } catch (Exception e) {
            progressService.fail(taskId, e.getMessage());
        }
    }


    private String extractAudio(String inputPath) throws Exception {
        String outPath = inputPath.replace(".mp4", ".wav");
        String cmd = String.format("ffmpeg -i %s -vn -acodec pcm_s16le -ar 16000 -ac 1 %s", inputPath, outPath);
        Runtime.getRuntime().exec(cmd).waitFor();
        return outPath;
    }


    private String callSpringAIWhisperAndTranslate(String audioPath) {
        // 偽代碼：可以集成 Spring AI Whisper + LLM 翻譯
        return "你好，歡迎來到 EasyDub。";
    }


    private String synthesizeAudio(String text) {
        // 偽代碼：調(diào)用 XTTS 合成中文音頻
        return "output/temp_tts.wav";
    }


    private String composeVideo(String originalVideo, String newAudio, String taskId) throws Exception {
        String output = baseDir + taskId + "_linly_dubbing.mp4";
        String cmd = String.format("ffmpeg -i %s -i %s -map 0:v -map 1:a -c:v copy -c:a aac %s",
                originalVideo, newAudio, output);
        Runtime.getRuntime().exec(cmd).waitFor();
        return output;
    }


    public String getProgress(String taskId) {
        return progressService.query(taskId);
    }


    public String getDownloadUrl(String taskId) {
        return progressService.getResultUrl(taskId);
    }
}

Redis 進(jìn)度服務(wù)封裝

package com.icoderoad.easydub.service;


import org.springframework.stereotype.Service;
import redis.clients.jedis.Jedis;


@Service
public class ProgressService {


    private final Jedis redis = new Jedis("localhost", 6379);


    public void init(String taskId) {
        redis.set(taskId, "開始處理...");
    }


    public void update(String taskId, String message) {
        redis.set(taskId, message);
    }


    public void complete(String taskId, String path) {
        redis.set(taskId + ":done", path);
        redis.set(taskId, "處理完成！");
    }


    public void fail(String taskId, String errorMsg) {
        redis.set(taskId, "失敗：" + errorMsg);
    }


    public String query(String taskId) {
        return redis.get(taskId);
    }


    public String getResultUrl(String taskId) {
        return redis.get(taskId + ":done");
    }
}

Web 前端 Thymeleaf + Bootstrap

index.html

<!DOCTYPE html>
<html xmlns:th="http://www.thymeleaf.org">
<head>
    <title>EasyDub 配音生成</title>
    <link rel="stylesheet" />
</head>
<body class="container mt-5">
<h2>?? EasyDub 配音生成平臺</h2>


<form id="uploadForm" enctype="multipart/form-data">
    <input type="file" class="form-control" name="file" required/>
    <button type="submit" class="btn btn-primary mt-2">上傳并開始處理</button>
</form>


<div class="mt-3" id="status" style="display:none;">
    <h5>進(jìn)度：<span id="progressMsg"></span></h5>
</div>


<div class="mt-3" id="download" style="display:none;">
    <a class="btn btn-success" id="downloadLink" href="#">下載結(jié)果視頻</a>
</div>


<script>
    document.getElementById('uploadForm').addEventListener('submit', function (e) {
        e.preventDefault();
        let formData = new FormData(this);
        fetch('/api/upload', {
            method: 'POST',
            body: formData
        }).then(res => res.text()).then(taskId => {
            document.getElementById("status").style.display = "block";
            pollProgress(taskId);
        });
    });


    function pollProgress(taskId) {
        let interval = setInterval(() => {
            fetch('/api/progress/' + taskId).then(res => res.text()).then(msg => {
                document.getElementById("progressMsg").innerText = msg;
                if (msg.includes("完成")) {
                    clearInterval(interval);
                    document.getElementById("download").style.display = "block";
                    fetch('/api/download/' + taskId).then(r => r.text()).then(url => {
                        document.getElementById("downloadLink").href = '/' + url;
                    });
                } else if (msg.includes("失敗")) {
                    clearInterval(interval);
                    alert("處理失敗：" + msg);
                }
            });
        }, 2000);
    }
</script>
</body>
</html>

本地 DEMO 流程

啟動 SpringBoot 應(yīng)用
瀏覽器打開 http://localhost:8080
上傳 original_video.mp4
等待進(jìn)度提示，后臺完成：

視頻 → 音頻提取 → Whisper識別 → 翻譯 → 合成配音 → 視頻合成

下載生成的 linly_dubbing.mp4

結(jié)語

通過整合 Spring Boot、Thymeleaf、Redis、FFmpeg 與 AI 模型接口（Whisper、XTTSv2 等），我們構(gòu)建了一個功能強大且易用的 EasyDub Web 配音系統(tǒng)，支持異步處理、狀態(tài)輪詢、數(shù)字人合成與完整視頻輸出。

責(zé)任編輯：武曉燕來源：路條編程