May 7, 2026 · Artificial Intelligence

OmniVoice: Open‑Source TTS Model Clones Voices in 600+ Languages with a Single Architecture

OmniVoice, an open‑source TTS system from Xiaomi AI Lab, uses a minimalist bidirectional Transformer and LLM‑enhanced pre‑training to synthesize high‑quality speech in over 600 languages, outperforming commercial systems while offering fine‑grained control and fully public code and models.

OmniVoiceOpen SourceTTS

0 likes · 8 min read

Alibaba Cloud Big Data AI Platform

Dec 16, 2025 · Artificial Intelligence

How CosyVoice 2.0 Cuts First‑Chunk Latency for High‑Fidelity Voice Cloning

CosyVoice 2.0, Alibaba DAMO Academy's next‑gen high‑fidelity speech synthesis model, introduces architecture decoupling, streaming generation, reference‑audio caching and dynamic load balancing to dramatically reduce first‑packet latency and improve real‑time factor while supporting multi‑language voice cloning.

AI model optimizationLow latencyStreaming Inference

0 likes · 9 min read

How CosyVoice 2.0 Cuts First‑Chunk Latency for High‑Fidelity Voice Cloning

OmniVoice: Open‑Source TTS Model Clones Voices in 600+ Languages with a Single Architecture

How CosyVoice 2.0 Cuts First‑Chunk Latency for High‑Fidelity Voice Cloning

How CosyVoice 2.0 Cuts First‑Chunk Latency for High‑Fidelity Voice Cloning