Fine-tuned XTTS-v2 ยท 10 speakers ยท Streaming ยท Code-switching โ Model not loaded โ check console
Speaker
Settings
Streaming = fast HTML5 player; Batch = waveform
Text Input
Lower = more stable, higher = more expressive
Higher reduces repeated sounds / breath artifacts
Streaming mode only (<1 slower, >1 faster)
๐ก Examples โ click any row
Upload a clean audio recording โ the model will synthesize in that voice.
๐ฟ Levantine: ูโ/ส/ ยท ุฌโ/ส/ ยท il- article ยท Dataset ยท Model