This is a specialized Text-to-Speech model dedicated exclusively to Akan (Asante Twi), specialized for the native voice of Speaker 17 (Female, Age 21).
from voxcpm import VoxCPM
import soundfile as sf
model = VoxCPM.from_pretrained('techolise/akan-twi-speaker17-tts')
wav = model.generate(text='Mema wo akye me nuanom! Wo ho te sɛn?')
sf.write('output.wav', wav, 16000)
model.safetensors 1.6 GBaudiovae.pth 301 MBspeaker17_master_scaled.ckpt 14 MBspeaker17_peak_lora.ckpt 14 MBspeaker17_lora_weights.ckpt 7 MB