F5-TTS
توثق هذه الصفحة من Soniqo نموذج F5-TTS كما هو منفذ في speech-swift / speech-core. روابط Hugging Face موجودة أدناه بعد ملاحظات الدمج.
الصفحة الداخلية أولا
بطاقات الصفحة الرئيسية وقوائم الوثائق تشير إلى هذه الصفحة أولا؛ وتبقى روابط النموذج والحزم داخلها.
لمحة سريعة
| النموذج | F5-TTS |
|---|---|
| الدور | Zero-shot voice cloning from a short reference plus transcript |
| Backend | MLX fp16 |
| الإخراج | 24 kHz mono waveform |
| اللغات | English and Mandarin, mixed EN/ZH supported |
| الرخصة | CC-BY-NC-4.0 weights; non-commercial bundle |
| الحالة | Ready through speech speak --engine f5 and the F5TTS Swift product |
| المصدر | SWivid F5-TTS |
| منتج Swift | F5TTS |
| CLI / runtime | speech speak --engine f5 |
الاستخدام
المقتطف أدناه يطابق API أو الأمر الحالي في speech-swift.
# Clone a voice from a ~10 s reference clip plus its transcript.
.build/release/speech speak "Hello from my cloned voice." \
--engine f5 \
--voice-sample reference.wav \
--f5-reference-text "Text spoken in the reference clip." \
-o cloned.wavروابط النموذج
ملاحظات التنفيذ
- 16 flow steps are the default (RTF 0.57 on an M5 Pro, lower is faster); --f5-steps 32 for maximum fidelity, 12 runs at RTF 0.43 — ASR roundtrips are word-identical across all three.
- --f5-reference-text is required: cloning conditions on the reference transcript, and quality tracks how cleanly the reference clip ends on a sentence boundary.
- Mandarin runs through a bundled pinyin lexicon with tone sandhi baked in at export time (99.2% token-exact against the upstream rjieba/pypinyin frontend), so Swift needs no external tokenizer.
- Deterministic per --f5-seed; --f5-speed scales pacing, CFG strength and sway sampling are exposed as flags; about 0.8 GB peak RSS makes it the smallest cloning engine in the package.