Views
No views yet
| Field | Value |
|---|---|
| Architecture | F5-TTS v1 Base (Flow Matching + DiT) |
| Base model | SWivid/F5-TTS |
| Training data | 439 clips / ~70 min |
| Sample rate | 24 kHz |
| Language | Arabic MSA + full Tashkeel |
| Final loss | 0.367 (20 epochs) |
| Author | Yahya (NightPrince) |
1from f5_tts.api import F5TTS
2from huggingface_hub import hf_hub_download
3
4ckpt = hf_hub_download("NightPrince/MSA-F5-TTS-Arabic", "model.pt")
5vocab = hf_hub_download("NightPrince/MSA-F5-TTS-Arabic", "vocab.txt")
6
7tts = F5TTS(model="F5TTS_v1_Base", ckpt_file=ckpt, vocab_file=vocab, device="cuda")
8
9wav, sr, _ = tts.infer(
10 ref_file="reference.wav",
11 ref_text="النص المرجعي",
12 gen_text="بِسْمِ اللَّهِ الرَّحْمٰنِ الرَّحِيمِ",
13 nfe_step=32
14)