Explorez et intégrez les meilleurs modèles d'IA via une API unique.
Tongyi Lab's offline text-to-speech model featuring 17 expressive human-like voices with low-latency, stable synthesis. Supports multiple languages and dialects, maintaining consistent timbre across languages. Adapts tone based on text semantics, handling complex content robustly
Vidu's text-to-audio generation model (model ID: audio1.0). Generates sound effects and background music from text prompts. Duration range: 2–10 seconds. Supports random seed configuration for reproducible outputs