 zhifu gaoandLauraGPT
|
842e62f875
|
docs: CPU benchmark vs whisper.cpp (Chinese ASR) (#2992)
* docs: add CPU benchmark vs whisper.cpp (FunASR is more accurate & faster on Chinese)
* docs(benchmark): clarify size column; add reproduction scripts (compute_cer.py + method README)
* docs(benchmark): finalize — canonical micro-CER+normalize_zh, full 184, fp32-ref/Q8-runtime dual values, whisper full set, transparency footnotes
---------
Co-authored-by: LauraGPT <[email protected]>
|
2026-06-19 22:26:14 +08:00 |
|