Search Results for "mandarin"

TAME LLM is an open-source initiative focused on building and releasing large language models optimized for Traditional Mandarin and the linguistic context of Taiwan. The project includes models such as Llama-3-Taiwan-70B, which are fine-tuned versions of large transformer architectures trained on extensive corpora containing both Traditional Mandarin and English text. These models are designed to support applications such as conversational AI, knowledge retrieval, and domain-specific reasoning in fields like manufacturing, law, healthcare, and electronics. ...

Chinese-XLNet is a Chinese language pre-trained model based on the XLNet architecture, providing an advanced foundation for natural language processing tasks in Mandarin and other Chinese dialects. Unlike traditional masked language modeling, XLNet uses a permutation language modeling objective that captures bidirectional context more effectively by training over all possible token orderings, yielding richer contextual representations. This model is trained on large-scale Chinese text datasets to learn linguistic patterns, long-range dependencies, and semantic nuance typical of Chinese writing, making it useful for tasks like text classification, question answering, named entity recognition, and language generation. ...

Chinese-LLaMA-Alpaca-3 is an open-source project that provides Mandarin-focused large language models based on Meta’s LLaMA-3 architecture, with both foundational and instruction-tuned variants to support high-quality Chinese natural language understanding and generation. It extends the original LLaMA models with expanded Chinese vocabularies and additional pretraining on Chinese corpora to improve semantic encoding and decoding specifically for Chinese text.

...It includes a large collection of “Kaldi-style” recipes for many datasets such as LJSpeech, LibriTTS, VCTK, JSUT, CMU Arctic, and multiple singing voice corpora in Japanese, Mandarin, Korean, and more. The project provides pre-trained models, Colab demos, and example configurations, allowing researchers to quickly evaluate vocoder quality or adapt models to new datasets.

This project has the language data for spel, the main new codebase is at: https://gitlab.com/liberit/pyac A computer programming language using human language syntax for human-to-human and human-to-computer communication with high precision, supporting many languages. Currently has alpha prototype support for analytic versions of the UN languages English, Mandarin Chinese, Spanish, Arabic, Russian and French as well as a bunch of others in addition to the core mwak language. The alpha IDE is at http://spel.sourceforge.net/src/web/spel.html (wait for it to finish loading before clicking "translate") Since it is early prototype, it's not easy to use, If you are interested, join the mailing list. ...

This project provides an implementation of duration high-order hidden Markov model (DHO-HMM) in Java. It is compactible with JDK 5 & 6. It was used in the author's research on speech recognition of Mandarin digits. There are some Chinese words in this project and I am afraid that I don't have enough time to translate to English recently.