Bash: Commands to Convert Multi Lines to Single Line

I have thousands on many lines in a text file. Some consist of 2 lines, some have 1 line, as seen below:

1
00:01:18,479 --> 00:01:20,514
Dr. Oppenheimer.

2
00:01:24,052 --> 00:01:25,653
As we begin, I believe you have

3
00:01:25,686 --> 00:01:27,688
a statement to read
into the record.

4
00:01:29,523 --> 00:01:30,725
Yes, Your Honor.

5
00:01:30,758 --> 00:01:32,861
We're
not judges, Doctor.

6
00:01:32,895 --> 00:01:33,929
Of course.

......so on and on...........

I want to know how to convert a sentence that only consists of 2 lines into one line so that the final result is like this:

1
00:01:18,479 --> 00:01:20,514
Dr. Oppenheimer.

2
00:01:24,052 --> 00:01:25,653
As we begin, I believe you have

3
00:01:25,686 --> 00:01:27,688
a statement to read into the record.

4
00:01:29,523 --> 00:01:30,725
Yes, Your Honor.

5
00:01:30,758 --> 00:01:32,861
We're not judges, Doctor.

6
00:01:32,895 --> 00:01:33,929
Of course.

......so on and on...........

>Solution :

Perl to the rescue!

perl -00 -lpe '$i = 0; s/\n/(++$i > 2) ? " " : "\n"/ge' input.txt

• -00 turns on the "paragraph mode" which reads the file in blocks separated by more than one newline;
• -l removes the final newlines from the input and adds them to output;
• -p reads the file block by block, prints the implicit variable after running the code;
• s/\n/.../g replaces each newline with the ... part;
• s///e interprets the replacement as code to evaluate;
• (++$i > 2) ? " " : "\n" is used to keep the first two newlines, replacing any other by a space. The variable $i serves as a line counter inside each block, it’s incremented on each newline, and once it has seen 2 of them, the ternary operator ?: starts returning a space instead of a newline.