12月1月2月3月4月5月6月7月8月9月10月11月
周一周三周五
一年内 264355 次贡献
更少更多
- frostbyte_neo 推送了仓库 github-featured/modelscope--funasr 的 docs/readme-benchmark-cer 分支2ab542dc94 docs(README): add Chinese CER to benchmark (canonical micro-avg, consistent with funasr.com)
- frostbyte_neo 推送了仓库 github-featured/modelscope--funasr 的 docs/fix-readme-quickstart-runnable 分支83748a09be docs(README): make the quickstart runnable (add missing model.generate() call + CPU note)
- frostbyte_neo 推送了仓库 github-featured/modelscope--funasr 的 docs/iic-namespace 分支5f0737a48d docs(README): use canonical iic/ ModelScope namespace (was legacy damo/)
- frostbyte_neo 推送了仓库 github-featured/modelscope--funasr 的 docs/llama-cpp-readme-section 分支3c7f00098a docs(README): promote llama.cpp/GGUF runtime to a first-class Deploy section
- frostbyte_neo 推送了仓库 github-featured/modelscope--funasr 的 docs/benchmarks-vs-whisper 分支比较 4 提交 »de25eb2ebc docs(benchmark): drop macro mention entirely (micro+normalize_zh is the only口径)33f0cfd8f4 docs(benchmark): finalize — canonical micro-CER+normalize_zh, full 184, fp32-ref/Q8-runtime dual values, whisper full set, transparency footnotes02bf66273e docs(benchmark): clarify size column; add reproduction scripts (compute_cer.py + method README)bd19cc63f4 docs: add CPU benchmark vs whisper.cpp (FunASR is more accurate & faster on Chinese)
- frostbyte_neo 推送了仓库 github-featured/modelscope--funasr 的 docs/build-note 分支d4142d40d2 docs: build note — copy funasr-common alongside the example (shared miniaudio audio loader)
- frostbyte_neo 推送了仓库 github-featured/modelscope--funasr 的 dev_xw 分支比较 7 提交 »33f2d469a6 [Optimization] fake scaling for fp16aafbf57cf9 [Optimization] fix bug18a119f057 [Optimization] fix bug4e45373bb2 [Optimization] add launch service scriptsa2757cb559 [Optimization] deploy torchscript model by triton