Files
diodiogod-TTS-Audio-Suite/docs/ENGINE_COMPARISON.md
T
diodiogod 6c87d67ed7 Add YAML-driven engine documentation table generator
- Created comprehensive YAML source of truth for all TTS engines
- Implemented Python generator for 3 documentation tables:
  * ENGINE_COMPARISON.md - Main engine overview with capabilities
  * LANGUAGE_SUPPORT.md - Detailed language support matrix
  * FEATURE_COMPARISON.md - Feature capabilities comparison
- YAML captures all engine metadata without information loss
- Tables auto-generated via scripts/generate_engine_tables.py
- Added flag emojis to language support table for easy scanning
- Speed performance fixes: F5-TTS marked as "Very Fast"
2026-01-28 10:50:10 -03:00

4.0 KiB
Raw Blame History

TTS Engines Reference Tables

Engine Comparison

Engine Models Size TTS SRT VC Special Features Languages
F5-TTS Base, v1, E2TTS + 8 lang models ~1.2GB each ✅ ✅ ❌ Targeted Word/Speech Editing, Speed control 🇺🇸🇩🇪🇪🇸🇫🇷🇮🇹🇯🇵🇧🇷🇵🇱🇮🇳🇹🇭
ChatterBox EN, DE×3, IT, FR, RU, HY, KA, JA, KO, NO ~4.3GB ✅ ✅ ✅ Expressiveness slider 🇺🇸🇩🇪🇫🇷🇮🇹🇯🇵🇰🇷🇷🇺🇳🇴🇦🇲🇬🇪
ChatterBox 23L v1, v2, Vietnamese (Viterbox) ~4.3GB ✅ ✅ ✅ 24 languages in single model, emotion tokens (v2 - doesn't work) 🇺🇸🇨🇳🇩🇪🇪🇸🇫🇷🇮🇹🇯🇵🇰🇷🇷🇺🇵🇹🇵🇱🇮🇳🇦🇪🇹🇷🇹🇭🇳🇴🇻🇳🇩🇰🇫🇮🇬🇷🇮🇱🇲🇾🇳🇱🇸🇪🇰🇪(+9)
VibeVoice 1.5B, 7B, Hindi-1.5B, Hindi-7B 5.4GB / 18GB ✅ ✅ ❌ 90-min long-form, Native 4-speaker, 4-bit quantization 🇺🇸🇨🇳🇮🇳
Higgs Audio 2 3B ~9GB ✅ ✅ ❌ 3 multi-speaker, CUDA graphs (55+ tokens/sec) 🇺🇸🇨🇳🇩🇪🇪🇸🇰🇷
IndexTTS-2 IndexTTS-2 ? ✅ ✅ ❌ Emotion Control: 8 vectors, Text as reference, Audio as reference 🇺🇸🇨🇳🇯🇵
CosyVoice3 0.5B, 0.5B-RL ~5.4GB ✅ ✅ ✅ Paralinguistic tags 🇺🇸🇨🇳🇯🇵🇰🇷
Qwen3-TTS 0.6B, 1.7B (CustomVoice/VoiceDesign/Base) ~3-6GB ✅ ✅ ❌ Voice design 🇺🇸🇨🇳🇩🇪🇪🇸🇫🇷🇮🇹🇯🇵🇰🇷🇷🇺🇵🇹
Step Audio EditX 3B LLM + CosyVoice ~7GB ✅ ✅ ❌ Second Pass Speech Editing Node: 14 emotions, 32 speaking styles, Paralinguistic effects 🇺🇸🇨🇳🇯🇵🇰🇷
RVC Community .pth 100-300MB ❌ ❌ ✅ Real-time VC, Pitch shift (±14), 6 HuBERT models, Language-independent 🇺🇸🇨🇳🇩🇪🇪🇸🇫🇷🇮🇹🇯🇵🇰🇷🇷🇺🇧🇷🇵🇱🇮🇳🇦🇪🇹🇷🇹🇭🇳🇴🇻🇳🇦🇲🇬🇪🇩🇰🇫🇮🇬🇷🇮🇱🇲🇾🇳🇱🇸🇪🇰🇪