Skip to content

Formerly KrillinAI, VideoTranslation and Dubbing Tool for 120+ Languages. Open-source AI workspace for creators, powered by Codex. Video downloader, video translation, create videos, images, audios, and edits with Agents in one place. For your Social Media like YouTube, TikTok, Instagram, etc.

feat: add text result

沈浩e2daa7cd655f
diff --git a/internal/service/subtitle_service.go b/internal/service/subtitle_service.go
index 3c00ba7b..554f8c9c 100644
--- a/internal/service/subtitle_service.go
+++ b/internal/service/subtitle_service.go
@@ -111,22 +111,25 @@ func (s Service) StartSubtitleTask(req dto.StartVideoSubtitleTaskReq) (*dto.Star
}
stepParam := types.SubtitleTaskStepParam{
- TaskId: taskId,
- TaskBasePath: taskBasePath,
- Link: req.Url,
- SubtitleResultType: resultType,
- EnableModalFilter: req.ModalFilter == types.SubtitleTaskModalFilterYes,
- EnableTts: req.Tts == types.SubtitleTaskTtsYes,
- TtsVoiceCode: ttsVoiceCode,
- VoiceCloneAudioUrl: voiceCloneAudioUrl,
- ReplaceWordsMap: replaceWordsMap,
- OriginLanguage: types.StandardLanguageName(req.OriginLanguage),
- TargetLanguage: types.StandardLanguageName(req.TargetLang),
- UserUILanguage: types.StandardLanguageName(req.Language),
- EmbedSubtitleVideoType: req.EmbedSubtitleVideoType,
- VerticalVideoMajorTitle: req.VerticalMajorTitle,
- VerticalVideoMinorTitle: req.VerticalMinorTitle,
- OriginLanguageWordOneLine: req.OriginLanguageWordOneLine,
+ TaskId: taskId,
+ TaskBasePath: taskBasePath,
+ Link: req.Url,
+ SubtitleResultType: resultType,
+ EnableModalFilter: req.ModalFilter == types.SubtitleTaskModalFilterYes,
+ EnableTts: req.Tts == types.SubtitleTaskTtsYes,
+ TtsVoiceCode: ttsVoiceCode,
+ VoiceCloneAudioUrl: voiceCloneAudioUrl,
+ ReplaceWordsMap: replaceWordsMap,
+ OriginLanguage: types.StandardLanguageName(req.OriginLanguage),
+ TargetLanguage: types.StandardLanguageName(req.TargetLang),
+ UserUILanguage: types.StandardLanguageName(req.Language),
+ EmbedSubtitleVideoType: req.EmbedSubtitleVideoType,
+ VerticalVideoMajorTitle: req.VerticalMajorTitle,
+ VerticalVideoMinorTitle: req.VerticalMinorTitle,
+ OriginLanguageWordOneLine: 12, // 默认值
+ }
+ if req.OriginLanguageWordOneLine != 0 {
+ stepParam.OriginLanguageWordOneLine = req.OriginLanguageWordOneLine
}
go func() {
defer func() {
@@ -651,7 +654,9 @@ func (s Service) splitSrt(ctx context.Context, stepParam *types.SubtitleTaskStep
log.GetLogger().Info("audioToSubtitle.splitSrt start", zap.Any("task id", stepParam.TaskId))
originLanguageSrtFilePath := filepath.Join(stepParam.TaskBasePath, types.SubtitleTaskOriginLanguageSrtFileName)
+ originLanguageTextFilePath := filepath.Join(stepParam.TaskBasePath, "output", types.SubtitleTaskOriginLanguageTextFileName)
targetLanguageSrtFilePath := filepath.Join(stepParam.TaskBasePath, types.SubtitleTaskTargetLanguageSrtFileName)
+ targetLanguageTextFilePath := filepath.Join(stepParam.TaskBasePath, "output", types.SubtitleTaskTargetLanguageTextFileName)
// 打开双语字幕文件
file, err := os.Open(stepParam.BilingualSrtFilePath)
if err != nil {
@@ -660,13 +665,21 @@ func (s Service) splitSrt(ctx context.Context, stepParam *types.SubtitleTaskStep
}
defer file.Close()
- // 打开输出文件
+ // 打开输出字幕和文稿文件
originLanguageSrtFile, err := os.Create(originLanguageSrtFilePath)
if err != nil {
log.GetLogger().Error("audioToSubtitle.splitSrt os.Create originLanguageSrtFile err", zap.Any("stepParam", stepParam), zap.Error(err))
return err
}
defer originLanguageSrtFile.Close()
+
+ originLanguageTextFile, err := os.Create(originLanguageTextFilePath)
+ if err != nil {
+ log.GetLogger().Error("audioToSubtitle.splitSrt os.Create originLanguageTextFile err", zap.Any("stepParam", stepParam), zap.Error(err))
+ return err
+ }
+ defer originLanguageTextFile.Close()
+
targetLanguageSrtFile, err := os.Create(targetLanguageSrtFilePath)
if err != nil {
log.GetLogger().Error("audioToSubtitle.splitSrt os.Create targetLanguageSrtFile err", zap.Any("stepParam", stepParam), zap.Error(err))
@@ -674,6 +687,12 @@ func (s Service) splitSrt(ctx context.Context, stepParam *types.SubtitleTaskStep
}
defer targetLanguageSrtFile.Close()
+ targetLanguageTextFile, err := os.Create(targetLanguageTextFilePath)
+ if err != nil {
+ log.GetLogger().Error("audioToSubtitle.splitSrt os.Create targetLanguageTextFile err", zap.Any("stepParam", stepParam), zap.Error(err))
+ }
+ defer targetLanguageTextFile.Close()
+
isTargetOnTop := stepParam.SubtitleResultType == types.SubtitleResultTypeBilingualTranslationOnTop
scanner := bufio.NewScanner(file)
@@ -684,7 +703,7 @@ func (s Service) splitSrt(ctx context.Context, stepParam *types.SubtitleTaskStep
// 空行代表一个字幕块的结束
if line == "" {
if len(block) > 0 {
- util.ProcessBlock(block, targetLanguageSrtFile, originLanguageSrtFile, isTargetOnTop)
+ util.ProcessBlock(block, targetLanguageSrtFile, targetLanguageTextFile, originLanguageSrtFile, originLanguageTextFile, isTargetOnTop)
block = nil
}
} else {
@@ -693,7 +712,7 @@ func (s Service) splitSrt(ctx context.Context, stepParam *types.SubtitleTaskStep
}
// 处理文件末尾的字幕块
if len(block) > 0 {
- util.ProcessBlock(block, targetLanguageSrtFile, originLanguageSrtFile, isTargetOnTop)
+ util.ProcessBlock(block, targetLanguageSrtFile, targetLanguageTextFile, originLanguageSrtFile, originLanguageTextFile, isTargetOnTop)
}
if err = scanner.Err(); err != nil {
@@ -1025,13 +1044,12 @@ func (s Service) generateTimestamps(taskId, basePath string, originLanguage type
// 生成短句子的英文字幕
var (
- originSentence string
- startWord types.Word
- endWord types.Word
- shortSentenceWordNum = originLanguageWordOneLine //控制单行英文的字数
+ originSentence string
+ startWord types.Word
+ endWord types.Word
)
- if len(sentenceWords) <= shortSentenceWordNum {
+ if len(sentenceWords) <= originLanguageWordOneLine {
shortOriginSrtMap[srtBlock.Index] = append(shortOriginSrtMap[srtBlock.Index], util.SrtBlock{
Index: srtBlock.Index,
Timestamp: fmt.Sprintf("%s --> %s", util.FormatTime(float32(sentenceTs.Start+tsOffset)), util.FormatTime(float32(sentenceTs.End+tsOffset))),
@@ -1040,19 +1058,19 @@ func (s Service) generateTimestamps(taskId, basePath string, originLanguage type
continue
}
- if len(sentenceWords) > 8 && len(sentenceWords) <= 2*shortSentenceWordNum {
- shortSentenceWordNum = len(sentenceWords)/2 + 1
- } else if len(sentenceWords) > 2*shortSentenceWordNum && len(sentenceWords) <= 3*shortSentenceWordNum {
- shortSentenceWordNum = len(sentenceWords)/3 + 1
- } else if len(sentenceWords) > 3*shortSentenceWordNum && len(sentenceWords) <= 4*shortSentenceWordNum {
- shortSentenceWordNum = len(sentenceWords)/4 + 1
- } else if len(sentenceWords) > 4*shortSentenceWordNum && len(sentenceWords) <= 5*shortSentenceWordNum {
- shortSentenceWordNum = len(sentenceWords)/5 + 1
+ if len(sentenceWords) > 8 && len(sentenceWords) <= 2*originLanguageWordOneLine {
+ originLanguageWordOneLine = len(sentenceWords)/2 + 1
+ } else if len(sentenceWords) > 2*originLanguageWordOneLine && len(sentenceWords) <= 3*originLanguageWordOneLine {
+ originLanguageWordOneLine = len(sentenceWords)/3 + 1
+ } else if len(sentenceWords) > 3*originLanguageWordOneLine && len(sentenceWords) <= 4*originLanguageWordOneLine {
+ originLanguageWordOneLine = len(sentenceWords)/4 + 1
+ } else if len(sentenceWords) > 4*originLanguageWordOneLine && len(sentenceWords) <= 5*originLanguageWordOneLine {
+ originLanguageWordOneLine = len(sentenceWords)/5 + 1
}
i := 1
for _, word := range sentenceWords {
- if i == 1 || i%(shortSentenceWordNum+1) == 0 {
+ if i == 1 || i%(originLanguageWordOneLine+1) == 0 {
startWord = word
if startWord.Start < endWord.End {
startWord.Start = endWord.End
@@ -1076,7 +1094,7 @@ func (s Service) generateTimestamps(taskId, basePath string, originLanguage type
endWord.End = sentenceTs.End
}
- if i%shortSentenceWordNum == 0 && i > 1 {
+ if i%originLanguageWordOneLine == 0 && i > 1 {
shortOriginSrtMap[srtBlock.Index] = append(shortOriginSrtMap[srtBlock.Index], util.SrtBlock{
Index: srtBlock.Index,
Timestamp: fmt.Sprintf("%s --> %s", util.FormatTime(float32(startWord.Start+tsOffset)), util.FormatTime(float32(endWord.End+tsOffset))),
@@ -1084,7 +1102,6 @@ func (s Service) generateTimestamps(taskId, basePath string, originLanguage type
})
originSentence = ""
}
-
i++
}
diff --git a/internal/types/subtitle_task.go b/internal/types/subtitle_task.go
index 6d5392b0..e0d955cf 100644
--- a/internal/types/subtitle_task.go
+++ b/internal/types/subtitle_task.go
@@ -131,7 +131,9 @@ const (
SubtitleTaskShortOriginMixedSrtFileName = "short_origin_mixed_srt.srt" //长中文+短英文
SubtitleTaskShortOriginSrtFileName = "short_origin_srt.srt" //短英文
SubtitleTaskOriginLanguageSrtFileName = "origin_language_srt.srt"
+ SubtitleTaskOriginLanguageTextFileName = "origin_language.txt"
SubtitleTaskTargetLanguageSrtFileName = "target_language_srt.srt"
+ SubtitleTaskTargetLanguageTextFileName = "target_language.txt"
SubtitleTaskStepParamGobPersistenceFileName = "step_param.gob"
SubtitleTaskTransferredVerticalVideoFileName = "transferred_vertical_video.mp4"
SubtitleTaskHorizontalEmbedVideoFileName = "horizontal_embed.mp4"
diff --git a/main.exe b/main.exe
deleted file mode 100644
index 4476f263..00000000
Binary files a/main.exe and /dev/null differ
diff --git a/pkg/util/subtitle.go b/pkg/util/subtitle.go
index 9d567666..57f7c8db 100644
--- a/pkg/util/subtitle.go
+++ b/pkg/util/subtitle.go
@@ -2,10 +2,8 @@ package util
import (
"bufio"
- "encoding/json"
"fmt"
"os"
- "os/exec"
"path/filepath"
"regexp"
"strconv"
@@ -13,18 +11,8 @@ import (
"unicode"
)
-// 判断是否包含中文字符
-func containsChinese(s string) bool {
- for _, r := range s {
- if unicode.Is(unicode.Han, r) {
- return true
- }
- }
- return false
-}
-
// 处理每一个字幕块
-func ProcessBlock(block []string, targetLanguageFile, originLanguageFile *os.File, isTargetOnTop bool) {
+func ProcessBlock(block []string, targetLanguageFile, targetLanguageTextFile, originLanguageFile, originLanguageTextFile *os.File, isTargetOnTop bool) {
var targetLines, originLines []string
// 匹配时间戳的正则表达式
timePattern := regexp.MustCompile(`\d{2}:\d{2}:\d{2},\d{3} --> \d{2}:\d{2}:\d{2},\d{3}`)
@@ -38,21 +26,25 @@ func ProcessBlock(block []string, targetLanguageFile, originLanguageFile *os.Fil
if len(targetLines) == 2 && len(originLines) == 2 { // 刚写完编号和时间戳,到了上方的文字行
if isTargetOnTop {
targetLines = append(targetLines, line)
+ targetLanguageTextFile.WriteString(line) // 文稿文件
} else {
originLines = append(originLines, line)
+ originLanguageTextFile.WriteString(line)
}
continue
}
// 到了下方的文字行
if isTargetOnTop {
originLines = append(originLines, line)
+ originLanguageTextFile.WriteString(line)
} else {
targetLines = append(targetLines, line)
+ targetLanguageTextFile.WriteString(line)
}
}
if len(targetLines) > 2 {
- // 写入中文文件
+ // 写入目标语言文件
for _, line := range targetLines {
targetLanguageFile.WriteString(line + "\n")
}
@@ -60,7 +52,7 @@ func ProcessBlock(block []string, targetLanguageFile, originLanguageFile *os.Fil
}
if len(originLines) > 2 {
- // 写入英文文件
+ // 写入源语言文件
for _, line := range originLines {
originLanguageFile.WriteString(line + "\n")
}
@@ -88,31 +80,6 @@ type ProbeData struct {
Format Format `json:"format"`
}
-// 获取 MP3 文件的时长(秒)
-func GetMP3Duration(filePath string) (uint32, error) {
- // 调用ffprobe获取音频文件的时长
- cmd := exec.Command("ffprobe", "-v", "error", "-show_entries", "format=duration", "-of", "json", filePath)
- output, err := cmd.Output()
- if err != nil {
- return 0, err
- }
-
- // 解析ffprobe返回的JSON数据
- var probeData ProbeData
- err = json.Unmarshal(output, &probeData)
- if err != nil {
- return 0, err
- }
-
- // 将时长字符串转换为浮点数
- duration, err := strconv.ParseFloat(probeData.Format.Duration, 64)
- if err != nil {
- return 0, err
- }
-
- return uint32(duration), nil
-}
-
type SrtBlock struct {
Index int
Timestamp string
diff --git a/static/index.html b/static/index.html
index 98d9f7ee..66f70786 100644
--- a/static/index.html
+++ b/static/index.html
@@ -371,6 +371,8 @@
</div>
<div id="downloadLinks" class="hidden download-links"></div>
</div>
+ <div id="tips" class="">
+ </div>
</div>
<script>
@@ -406,6 +408,7 @@
const subtitleTextInputs = document.getElementById("subtitle-text-inputs");
const verticalMajorTitle = document.getElementById("subtitle-text-1");
const verticalMinorTitle = document.getElementById("subtitle-text-2");
+ const tips = document.getElementById("tips");
const API_SUBMIT_URL = "/api/capability/subtitleTask";
const API_PROGRESS_URL = "/api/capability/subtitleTask";
@@ -670,6 +673,7 @@
downloadLinks.innerHTML = ''; // 清空之前的下载链接
if (taskId) {
+ executeButton.disabled = true;
// 显示任务进度区域
progressSection.classList.remove('hidden');
await pollTaskProgress(taskId);
@@ -712,6 +716,7 @@
});
console.log('progress response---:', response);
if (+response.error !== 0 && +response.error !== 200) {
+ executeButton.disabled = false;
throw new Error(response.msg || '查询进度失败');
}
const data = response.data || {};
@@ -720,12 +725,16 @@
progressBar.textContent = progress + '%';
if (progress >= 100) {
displayDownloadLinks(data.subtitle_info || [], data.speech_download_url || '');
+ executeButton.disabled = false;
+ // 显示output路径说明
+ tips.textContent = "若需要查看合成的视频或者文字稿,请到软件目录下的/tasks/"+data.task_id+"/output 目录下查看。";
return;
}
await new Promise(resolve => setTimeout(resolve, 5000));
}
} catch (error) {
alert(error);
+ executeButton.disabled = false;
console.error('查询进度失败:', error);
}
}