Objective T1a glottic cancer can be treated with transoral laser microsurgery (TLM) or radiation therapy (RT). Dysphonia is the major disadvantage of TLM, and preservation of voice appears to be the best advantage of RT compared to TLM. Studies on voice outcomes of both options gave conflicting results, but there is a tendency toward better voice outcome after TLM compared to the past. Study Design Nonrandomized retrospective cohort study. Setting Tertiary referral center. Methods In total, 172 patients with a cancer lesion involving more than two-thirds of 1 membranous vocal fold underwent cordectomy type I (n = 56) (C1 group) or type II (n = 59) (C2 group) or RT (n = 57) (RT group). GRBASI (grade, roughness, breathiness, asthenia, strain, instability), videolaryngostroboscopy, Voice Handicap Index–30, acoustic analysis including F0, jitter, shimmer, noise to harmonic ratio, cepstral peak prominence, and cepstral spectral index of dysphonia using running speech, and aerodynamic analysis were performed before treatment and 6 and 24 months after treatment. Results Study groups did not differ significantly on pretreatment voice outcomes ( P > .05). The RT group had significantly better voice outcomes at 6 months posttreatment compared to the C1 and C2 groups ( P < .05). The C1 group had significantly better voice outcomes at 6 months posttreatment compared to the C2 group ( P < .05). The C1 group had significantly better voice outcomes at 24 months posttreatment compared to the RT and C2 groups ( P < .05). The RT and C2 groups did not differ significantly at 24 months posttreatment ( P > .05). Conclusion Voice outcomes after C2 are equal to RT. C1 has better voice outcomes than RT. C1 and C2 can be the treatment of choice for large T1a glottic cancers.