In contrast to all the “it depends” and “it’s subjective” comments out here, I’ll point out 4 qualities that make sound high-fidelity, IMO.

1. Full frequency coverage: the system needs to cover at least 20 Hz to 20 KHz. While it is true that most people can’t hear under 30 Hz and over 16 KHz depending on age, harmonics matter and has significant effect on “coloration” of sound. Some will point out low distortion, which is true, but some harmonic distortion makes some systems (especially tubes) better than others. Zero distortion sounds very clinical and may take the sweetness away from listening. I’d recommend Yo-Yo Ma’s Legend of Herlen (from Silk Road Journeys) to feel the sharp ringing of the bells and the deep, bassy notes of drums.

2. Transparency: the system should reproduce the music faithfully, to the source. This means the sound is free of clipping and distortion. If you listen to music on airline headphones or car stereos, for example, it removes all fine detail and makes the sound “hollow”. Some systems enhance the bass/treble/midrange (my pet peeve with most live concerts - too much bass, so it feels like “party music”), rendering sound “not the way the artist intended”. When you can render music faithfully, you can hear fine detail, like guitar strings vibrating or a tabla being struck by a finger. To see examples, try listening to Nils Lofgren’s Keith don’t go from Acoustic live, or Zakir Hussain and Rajesh Chaurasia’s concert from EtnoKrakow. You can feel the steel strings of the guitar, the wooden quality of the bansuri and the fine sounds of the tabla on good speakers. Another example is the start of Wish you were here by Pink Floyd - it starts off sounding “recorded”, but you can tell when the “real” guitar kicks in because it feels so natural.

3. Soundstage: you should be able to isolate every unique source on “stage”, meaning the placement of singer and instruments (to the right/left) should be apparent, even when there are more than 2-3 sources). This also applies to depth in some cases - hifi systems sound richly layered, and you can tell if one instrument is placed behind a singer / another instrument. Height of the sound is another - and all of this happens not just because of R/L speaker loudness, but also phase. There’s also an aspect of “focus” - sounds have precisely defined outlines or boundaries in the soundstage. I’d recommended listening to Imagine by Rachel Z, and Spanish Harlem by Rebecca Pidgeon to observe this quality.

4. Dynamic Range: quiet sounds should be quiet, and loud sounds should be loud. Music compression often involves “normalizing” volume. When you hear a true hifi recording, you’ll see that you’re hearing ambient noise if you pay attention (Roger Waters clearing his throat, drawing a deep breath before singing Wish you were here), but it is quiet and barely noticeable when you’re not paying attention. Another great track is Novocaine by Amber Rubarth.

Also, while the components are important to resolve the fine detail, the music recording needs to be high-resolution (the mic, mic preamp, no compression, storage format, etc.,) from the source. Hifi components cannot “manufacture” high-fidelity detail that’s simply not present in the source, so having an analog source like a record player, or DSD files and a DSD compatible DAC is important (and also why this hobby is a rabbit hole).

I’d highly recommend trying Chesky records and samplers, they make some fantastic recordings and add helpful narration to train your ears and notice qualities in music. Enjoy!