Most people hate to hear recordings of their own speech. "Oh God, do I really sound like that?" Many people think that their voice is "bad", which should be surprising because almost nobody has an objectively bad voice. But many people use their voices unskillfully, which can be improved.
You might hate the sound of your voice because it doesn't sound like you think it does. You're accustomed to hearing your voice through your skull, which boosts the lower frequencies. Consequently, your voice seems deeper and more dynamic than it actually is. When you listen to a recording, your voice sounds higher and less harmonically rich than you're used to.
There are also cultural beliefs about voice that are best dispensed with. For example, you may think that men should have deep voices, and if you are a man, you may feel that your voice is too high. First of all, it probably isn't, actually – it's just that when you hear a recording of yourself, it's higher than you think it is, and the contrast is jarring. Second, nobody thinks any less of Elijah Wood, for example, for having a relatively high-register speaking voice. The received wisdom is simply wrong.
And third: if you're using your voice dynamically, your pitch will move up and down the register. In fact, it's more common for men to speak too low, as men sometimes aim for an exaggerated deep voice at the low end of their range. That degree of depth doesn't suit their physiology and tends to sound flat and monotone.
The timbre (pronounced "tamber") or "tone quality" or "texture" of a sound are all the aspects of the sound that don't specifically relate to its pitch. Timbre is the element of sound that lets you tell the difference between a note played on a cello and an electric guitar.
While most of your voice's timbre is caused by anatomy, you can control some of its aspects. You can actively change where your voice resonates; you can speak in "chest voice", "throat voice" or "head voice." You can make this change with or without changing the pitch. Aim for a voice between chest and head, which captures some chest resonance without sacrificing the higher head harmonics. Most people enjoy this timbre more than either extreme.
To explore the resonance of your voice, practice making "siren" sounds. Start low in your chest and rise up into the nasal range as your pitch increases. Consider where you usually find yourself speaking along this scale. Does your resonance change when you're in a loud and crowded environment?
Whether in conversation or speaking before an audience, you will come across as flat and monotone if you stick with one timbre and pitch, In contrast, varying your voice makes you sound excited and makes people want to listen.
Another reason you might hate to hear yourself speaking is that you suddenly become aware of filler words, stammers, and other tics. You may have been told that you should never use these words. This is nonsense.
Spoken language is different from written text, and filler words ("um", "uh", "eh", "like", "hmm", "y'know") are tools used in speech for specific purposes. In the same way you learned language before grammar, you learned to use these sounds to inject shadings and nuance into your sentences.
"Yeah, like, I don't know if that's, uh, quite right," is a natural way to talk. Depending on the context, these filler words might convey humor, diffidence, or condescension.
Exercise. Read the sentence "Yeah, like, I don't know if that's, uh, quite right." in each of these three styles:
Pay attention to the way that the filler words ("Yeah", "like" and "uh") change with each reading. Then read it without the filler words ("I don't know if that's quite right.") and observe how much texture and nuance is lost. Filler words are not just filler. They are tools for communication like any other part of speech — but you still need to know how to use them appropriately.
Colloquialisms are another part of speech that require contextual awareness. Informal contractions such as "whatcha", "didja", "gotta" are improper to use in public speaking or professional conversations. However, these words are common in casual conversation. It's best to use informal speech patterns while retaining an awareness of your current context.
Benign conversational stammering is another vocal phenomenon that is unfairly maligned. "Yeah, I-I-uh-I-don't know if that's quite right." Benign stammering tends to happen when you haven't decided how to phrase the sentence before opening your mouth. Other uses of stammering include resetting your thoughts or speech cadence without pausing. These uses are fine.
In a conversation involving three or more people, it is often better to properly time when you start talking and stammer slightly. Waiting until your next utterance is perfectly formed in your head is an excellent recipe for never getting a chance to speak. And if you feel like you shouldn't be stammering, and you think stammering is a sign of being bad at speaking, then you will sound uncertain every time you do this (totally natural) speech behavior. Listen to the most confident human being in the world, Bill Burr, stammer and use filler words during his comedy bit.
There is an argument that filler words and stammering can be a distracting crutch. But, usually, the problem is not the use of filler words per se, but the lack of familiarity with the subject matter. Eliminating filler words from prepared speeches requires practice. If you are giving a presentation, it can help to write out exactly what you intend to say. Practice reading it aloud several times and the patterns of speech will stick, even if the exact sentences don't.
If you have gone through life up to this point disliking your own voice, consider that you might be the only person in the whole world who feels that way about your specific voice. Your energies would be much better spent studying how your voice and speech patterns come across outside of your own skull. Recording yourself is the most powerful tool to accomplish this. Short of that, you can take whatever natural voice you have and make it more engaging by learning to dynamically vary pitch and resonance. Finally, filler words, colloquialisms, and other non-word sounds should be confidently used on purpose, rather than accidentally with embarrassment.