Trained on a 12 min 40khz lossless dataset with RMVPE-Titan with batch size 8.
Model: https://huggingface.co/Albinator/John-Lennon-Talking-2/resolve/main/JohnLennonTalking2_e450_s14850.zip
Dataset Sources:
A Hard Day's Night Movie (23 Seconds with Adobe Podcast)
Help Movie (30 Seconds with Adobe Podcast)
Magical Mystery Tour Movie (2,09 minutes)
Live at BBC Albums (59 Seconds)
Christmas Albums (3,33 minutes)
Bootlegs (4,21 minutes)
Samples: #1269057582975287478 message https://www.weights.gg/sv/models/clzdak3qt0462sody5tigqc4f
Dataset: https://huggingface.co/datasets/Albinator/John-Lennon-Talking-2
If you use the dataset make sure to give me credit 🙂
It's not necessary but feel free to tag me on YouTube @Albinator_ so i can hear the stuff that you make 🙂
More Beatles speech models:
https://discord.com/channels/1159260121998827560/1176972941376901221
https://discord.com/channels/1159260121998827560/1269651775456153723
https://discord.com/channels/1159260121998827560/1177972935407968287
https://discord.com/channels/1159260121998827560/1199559895406628904
https://discord.com/channels/1159260121998827560/1249084990520557600
https://discord.com/channels/1159260121998827560/1270468160973443162
