[文本处理] [已解决]求联想语音带时间轴的txt转srt的批处理

#@&cls&powershell "type %~s0|out-string|iex"&pause&exit
sc a.srt $(
gc a.txt|%{
if($_ -replace '\.',',' -match '\d\d:\d\d:\d\d,\d{3}'){
$a+=@($matches[0])
if($a.count -eq 3){$i++;$i;'{0} --> {1}' -f $a[0],$a[2];$a[1],'';$a=@($a[2])}
}else{$a+=@($_)}
}
$i++;$i;$a
)
复制代码

<#*,:&cls
@echo off
cd /d "%~dp0"
powershell -NoProfile -ExecutionPolicy RemoteSigned -Command ". ([ScriptBlock]::Create((Get-Content -LiteralPath \"%~0\" -ReadCount 0 | Out-String ))) "
pause
exit /b
#>
# 功能:当前目录txt转换成srt的格式，我知道缺少每句话的结束时间，结束时间就用下一句的开始时间就行：

function GenSrt {
  param(
    [string]$Source,
    [string]$Output
  )
  $re = New-Object System.Text.RegularExpressions.Regex -ArgumentList @('^[^*\n]+\*+\s*(?<begin1>\d{2}:\d{2}:\d{2})\.(?<begin2>\d{3}).*$\n(?<talk>[\s\S]*?)(?=^[^*\n]+\*+\s*(?<end1>\d{2}:\d{2}:\d{2})\.(?<end2>\d{3})|\z)', 'Multiline,Compiled')
  $evaluator = [System.Text.RegularExpressions.MatchEvaluator] {
    param([System.Text.RegularExpressions.Match]$m)
    $Script:ctr++
    # Write-Host ($m.Groups['talk'].Value) -ForegroundColor Green
    if ($m.Groups['end1'].Success) {
      $Script:format -f $Script:ctr, $m.Groups['begin1'].Value, $m.Groups['begin2'].Value, $m.Groups['end1'].Value, $m.Groups['end2'].Value, $m.Groups['talk'].Value
    } else {
      $Script:format -f $Script:ctr, $m.Groups['begin1'].Value, $m.Groups['begin2'].Value, $m.Groups['begin1'].Value, $m.Groups['begin2'].Value, $m.Groups['talk'].Value
    }
  }
  $Script:ctr = 0
  [System.IO.File]::WriteAllText($Output, ($re.Replace( [System.IO.File]::ReadAllText($Source, $encoding), $evaluator )), $encoding)
}
$encoding = [System.Text.Encoding]::UTF8
$Script:format = "{0}`r`n{1},{2} --> {3},{4}`r`n{5}`r`n"
Get-ChildItem -Path "*.txt" | ForEach-Object {
  if (-not $_.PSIsContainer) {
    $dstfile = $_.BaseName + ".srt"
    Write-Host "$($_.Name) --> $dstfile"
    GenSrt -Source $_ -Output $dstfile
  }
}
复制代码

#@&cls&powershell "type %~s0|out-string|iex"&pause&exit
foreach($j in dir *.txt){
sc -enc utf8 $($j.basename+'.srt') $(
gc -enc utf8 $j|%{
if($_ -replace '\.',',' -match '\d\d:\d\d:\d\d,\d{3}'){
$a+=@($matches[0])
if($a.count -gt 1){$i++;$i;'{0} --> {1}' -f $a[0],$a[-1];$a[1..$($a.count-2)],'';$a=@($a[-1])}
}else{$a+=@($_)}
}
$i++;$i;$a;rv a,i
)
}
复制代码