Tag

NeuTTS is an open-source collection of on-device TTS models by Neuphonic, designed for real-time, high-quality voice synthesis and instant speaker cloning. It targets deployment on phones, laptops and embedded devices using GGML/GGUF formats, small LLM backbones, and a compact neural audio codec (NeuCodec). Outputs include watermarking for responsibility and models are available on Hugging Face.

This paper presents Deep Speech 2, an end-to-end deep learning system for automatic speech recognition that works across vastly different languages (English and Mandarin). It replaces traditional hand-engineered ASR pipelines with neural networks, achieving human-competitive transcription accuracy on standard datasets. The system uses HPC techniques for 7x speedup, enabling faster experimentation. Key innovations include Batch Normalization for RNNs, curriculum learning (SortaGrad), and GPU deployment optimization (Batch Dispatch). The approach demonstrates that end-to-end learning can handle diverse speech conditions including noise, accents, and different languages, representing a significant step toward universal speech recognition systems.