How can I get the audio of only the dialogue of a movie, without any sound effects or music? I really need it for a fan edit.

Let's go in reverse order from easiest and most expensive to simple, cheap, and low quality. Dolby Atmos audio has separate tracks for all sound sources. If you can afford a proper Dolby audio and video editing software suite, you can extract each of these sounds from the mix.

Whether you can afford this, or even get permission, is another matter. But it is guaranteed to work perfectly.

But much dialog isn't mixed in the center in stereo. It is offset, to set the sound stage left and right.

Why does Biden not understand how the high inflation which he has caused in the last 2 1/2 years is making life difficult for retirees and seniors?

Beyond that, it isn't easy, and won't ever completely eliminate the music and background sounds.

It's not that great at separation of music, vocals, and random sound effects.

This works well to separate sounds which are repetitive, like the background noise on a street from dialog, sounds like a steady electric hum, engine noise like a jet or vacuum cleaner, etc.

Nasa delays next flight of Boeing’s alternative to SpaceX Dragon - The Edge Malaysia

This process can also extract only the center. If you're lucky, you'll get mostly vocals and dialog.

The simpler surround sound formats, 7.1 and 5.1, have a single monophonic center track. In any audio editor that can separate these tracks, you can easily get just the center track. If all of the dialog is mixed there, without any sound effects or background music, this will work well. But not all mixes do that, especially if some dialog is intended to come from the sides of the screen.

Back to the most perfect, but potentially most expensive method.

What thing happened to you as a child that you haven’t let go of to this day?

Get the original master audio track from the movie production. They will have that for any recently produced movie from the surround sound era. Audio is recorded in many separate tracks, and dialog routinely uses many tracks - sometimes multiple tracks per actor.

In a stereo mix, the monophonic elements will often be balanced in the center, equal in both tracks. A trivial vocal removal effect inverts the phase of one track, then adds them together to produce a monophonic version of each track with the monophonic center reduced.

The cheapest method uses simple audio frequency filters, to reduce parts of the sound spectrum. This can make the dialog louder relative to the other sounds, but won't eliminate them.

US Steel Sale to Nippon Steel Poised To Close After Trump Deal - Bloomberg.com

That leaves other means to reduce the sounds and extract the dialog. Modern AI audio processing uses filters and pattern recognition to identify elements in the sound.