仅考虑单文件单说话人识别，导出 TXT 格式为

file: out.mp3
***********************************result**************************************
[0:0.000,1:0.220]  段落1
[1:0.220,1:13.320]  段落2
[1:16.080,1:18.440]  段落3

JavaScript网页版

注：程序仅在本地浏览器中执行。

C++ 转换程序样例

D:\qcloud_srt>srt.exe in.txt out.srt

#include <stdio.h>
#include <string.h>
#include <stdlib.h>

void convert_time(const char* input_time, char* output_time) {
    int minutes;
    float seconds;
    sscanf(input_time, "%d:%f", &minutes, &seconds);
    float total_seconds = minutes * 60 + seconds;
    int total_milliseconds = (int)(total_seconds * 1000 + 0.5);

    int hours = total_milliseconds / (3600 * 1000);
    int mins = (total_milliseconds % (3600 * 1000)) / (60 * 1000);
    int secs = (total_milliseconds % (60 * 1000)) / 1000;
    int milliseconds = total_milliseconds % 1000;

    sprintf(output_time, "%02d:%02d:%02d,%03d", hours, mins, secs, milliseconds);
}

int main(int argc, char * argv[]) {
    char infile[512], outfile[512];
    if (argc < 2) {
        printf("error: no input file\nusage: infile outfile\n");
        return 0;
    }
    strcpy(infile, argv[1]);
    if (argc == 3) strcpy(outfile, argv[2]);
    else strcpy(outfile, strcat(argv[1], ".srt"));

    FILE * fin, * fout;
    fin = fopen(infile, "r");
    if (!fin) {
        printf("error: cannot open infile\n");
        return 0;
    }
    fout = fopen(outfile, "w");
    if (!fout) {
        printf("error: cannot write outfile\n");
        fclose(fin);
        return 0;
    }

    char str[5120], str1[128], str2[128], str3[5120];
    int row = 1;
    while (fgets(str, 5120, fin)) {
        if (str[0] != '[') continue;
        int ret = sscanf(str, "[%[^,],%[^]]]  %[^\n]", str1, str2, str3);
        if (ret < 3) continue;

        char time1[16], time2[16];
        convert_time(str1, time1);
        convert_time(str2, time2);

        fprintf(fout, "%d\n%s --> %s\n%s\n\n", row++, time1, time2, str3);
    }
    fclose(fin);
    fclose(fout);

    return 0;
}