Karaoke feature does not work on windows

[edery-ai]

I tried to execute the command contained in the generated .wts file by converting the .wts into a .bat , but I received an error indicating that the system could not find the specified path.
I then attempted to modify the function in cli.cpp, between lines 767 and 886, but I believe I do not fully understand how the function works.

`static bool output_wts(struct whisper_context * ctx, std::ofstream & fout, const whisper_params & params, std::vector<std::vector> pcmf32s, const char * fname_inp, float t_sec, const char * fname_out) {
static const char * font = params.font_path.c_str();

std::ifstream fin(font);
if (!fin.is_open()) {
    fprintf(stderr, "%s: font not found at '%s', please specify a monospace font with -fp\n", __func__, font);
    return false;
}

fout << "#!/bin/bash" << "\n";
fout << "\n";

fout << "ffmpeg -i " << fname_inp << " -f lavfi -i color=size=1200x120:duration=" << t_sec << ":rate=25:color=black -vf \"";

for (int i = 0; i < whisper_full_n_segments(ctx); i++) {
    const int64_t t0 = whisper_full_get_segment_t0(ctx, i);
    const int64_t t1 = whisper_full_get_segment_t1(ctx, i);

    const int n = whisper_full_n_tokens(ctx, i);

    std::vector<whisper_token_data> tokens(n);
    for (int j = 0; j < n; ++j) {
        tokens[j] = whisper_full_get_token_data(ctx, i, j);
    }

    if (i > 0) {
        fout << ",";
    }

    // background text
    fout << "drawtext=fontfile='" << font << "':fontsize=24:fontcolor=gray:x=(w-text_w)/2:y=h/2:text='':enable='between(t," << t0/100.0 << "," << t0/100.0 << ")'";

    bool is_first = true;
    std::string speaker = "";

    if (params.diarize && pcmf32s.size() == 2) {
        speaker = estimate_diarization_speaker(pcmf32s, t0, t1);
    }

    for (int j = 0; j < n; ++j) {
        const auto & token = tokens[j];

        if (tokens[j].id >= whisper_token_eot(ctx)) {
            continue;
        }

        std::string txt_bg = "";
        std::string txt_fg = ""; // highlight token
        std::string txt_ul = ""; // underline

        if (params.diarize && pcmf32s.size() == 2) {
            txt_bg = speaker;
            txt_fg = speaker;
            txt_ul = "\\ \\ \\ \\ \\ \\ \\ \\ \\ \\ \\ ";
        }

        txt_bg.append("> ");
        txt_fg.append("> ");
        txt_ul.append("\\ \\ ");

        {
            for (int k = 0; k < n; ++k) {
                const auto & token2 = tokens[k];

                if (tokens[k].id >= whisper_token_eot(ctx)) {
                    continue;
                }

                const std::string txt = whisper_token_to_str(ctx, token2.id);

                txt_bg += txt;

                if (k == j) {
                    for (int l = 0; l < (int) txt.size(); ++l) {
                        txt_fg += txt[l];
                        txt_ul += "_";
                    }
                    txt_fg += "|";
                } else {
                    for (int l = 0; l < (int) txt.size(); ++l) {
                        txt_fg += "\\ ";
                        txt_ul += "\\ ";
                    }
                }
            }

            ::replace_all(txt_bg, "'", "\u2019");
            ::replace_all(txt_bg, "\"", "\\\"");
            ::replace_all(txt_fg, "'", "\u2019");
            ::replace_all(txt_fg, "\"", "\\\"");
        }

        if (is_first) {
            // background text
            fout << ",drawtext=fontfile='" << font << "':fontsize=24:fontcolor=gray:x=(w-text_w)/2:y=h/2:text='" << txt_bg << "':enable='between(t," << t0/100.0 << "," << t1/100.0 << ")'";
            is_first = false;
        }

        // foreground text
        fout << ",drawtext=fontfile='" << font << "':fontsize=24:fontcolor=lightgreen:x=(w-text_w)/2+8:y=h/2:text='" << txt_fg << "':enable='between(t," << token.t0/100.0 << "," << token.t1/100.0 << ")'";

        // underline
        fout << ",drawtext=fontfile='" << font << "':fontsize=24:fontcolor=lightgreen:x=(w-text_w)/2+8:y=h/2+16:text='" << txt_ul << "':enable='between(t," << token.t0/100.0 << "," << token.t1/100.0 << ")'";
    }
}

fout << "\" -c:v libx264 -pix_fmt yuv420p -y " << fname_inp << ".mp4" << "\n";

fout << "\n\n";
fout << "echo \"Your video has been saved to " << fname_inp << ".mp4\"" << "\n";
fout << "\n";
fout << "echo \"  ffplay " << fname_inp << ".mp4\"\n";
fout << "\n";

fout.close();

fprintf(stderr, "# %s: run 'source %s' to generate karaoke video\n", __func__, fname_out);

return true;

}
`