15 projects
kaldingram
Train and entropy-prune ARPA n-gram language models
rnnlm
RNN language model training, export, and perplexity toolkit
atdataset
Audio text dataset for PyTorch training based on webdataset.
ms2hf
Sync models and datasets between ModelScope and Hugging Face one file at a time.
zipformer
zipformer: A faster and better encoder for ASR
simple-sentencepiece
A simple sentencepiece encoder and decoder without any dependency.
nndiagnostics
Lightweight PyTorch tensor diagnostics hooks for training loops
yapformer
yapformer: Yet Another zaPFORMER.
plrun
Run shell commands in parallel.
zipvoice
ZipVoice: Fast and High-Quality Zero-Shot Text-to-Speech with Flow Matching
zapformer
zapformer
cppinyin
A simple and fast pinyin encoder and decoder.
fasttextsearch
Some fast-ish algorithms for batch text search in moderate-sized collections, intended for data cleanup.
fast-rnnt
Fast and memory-efficient RNN-T loss.
simple-wer
A simple commandline tool for calculating WERs.
