The first model got the other voices out, but my own voice still sounded like a phone

Free speech enhancement. Upload an audio or video file of one person and get that voice back regenerated clean, without the noise or the room. No account.