This is your work, valued
Was vorbei ist, ist vorbei.
kaldifeat. Kaldi-compatible online & offline feature extraction with PyTorch, supporting CUDA, batch processing, chunk processing, and autograd - Provide C++ & Python API
215onnxruntime-libs. onnxruntime pre-compiled libs
197kaldi-native-fbank. Kaldi-compatible online fbank extractor without external dependencies
152code-for-Lucas-Kanade-20-Years-On. C
74optimized_transducer. Memory efficient transducer loss computation
70transducer-loss-benchmarking. Python
67OpenCNN. An Open Convolutional Neural Network Framework in C++ From Scratch
67PortAudioSharp2. C# binding for portaudio supporting Linux, macOS, Windows, iOS
51kaldilm. Python wrapper for kaldi's arpa2fst
38onnxruntime-build. A build project for ONNX Runtime
30kaldi-hmm-gmm. C++
28sherpa-onnx. C++
28kaldi_native_io. python wrapper for kaldi's native I/O
27ppm. PatchMatch algorithms for motion estimation and stereo reconstruction
17naudiodon2. Node.js stream bindings for PortAudio
11icefall. Python
11vits_chinese. vits chinese, tts chinese, tts mandarin 史上训练最简单,音质最好的语音合成系统
9Operating-Systems-Three-Easy-Pieces-all-in-one-pdf. Operating Systems: Three Easy Pieces All in One pdf
8blas-lapack-in-c-cplusplus. Examples of using BLAS and Lapack in C and C++
7portaudio-go. Go bindings for the PortAudio audio I/O library
7vits-cantonese. VITS implementation of Japanese, Chinese, Korean, Sanskrit and Thai
6sr-data. Speaker identification test data for sherpa-onnx
5k2_decoding_benchmark. Python
4ncnn. ncnn is a high-performance neural network inference framework optimized for the mobile platform
4sherpa. Speech-to-text server framework with next-gen Kaldi
4kfj-vim. My vim settings.
3sherpa-ncnn. C++
2tts-dataset-creation. Create dataset for tts
2icefall-asr-conformer-ctc-bpe-500.
2Matcha-TTS. [ICASSP 2024] 🍵 Matcha-TTS: A fast TTS architecture with conditional flow matching
1csukuangfj.
1memory-leak-example. CMake
1subllm. This repository is the official implementation of the ECAI 2024 conference paper SUBLLM: A Novel Efficient Architecture with Token Sequence Subsampling for LLM
1mecat. Python
1cppjieba. "结巴"中文分词的C++版本
1piper. A fast, local neural text to speech system
1MeloTTS. High-quality multi-lingual text-to-speech library by MyShell.ai. Support English, Spanish, French, Chinese, Japanese and Korean.
1