llama.cpp 发布: b11190

来源摘要

mtmd: fix mel preprocessor in LFM2 audio (#29403) which resulted in different greedy transcripts for 4.5% of English and 6.5% of Japanese test utterances. In Japanese, some differences changed entire words. This change: * uses `log(x + 2^-24)` instead of clamping to the log floor * uses a symmetric Hann window, equivalent to `torch.hann_window(periodic=False)` * adds the normalization epsilon to the standard deviation instead of inside the square root Only the `lfm2a` preprocessor opts into these behaviors. Other audio preprocessors are unchanged. Tested on top of 84e76d8 using `llama-server` with CUDA and `temperature=0`, compared against http://github.com/Liquid4All/liquid-audio fp32. Test set: * 200 LibriSpeech `test-clean` utterances (EN) * 200 Common Voice `ja` test utterances (JP) * identical 16 kHz audio passed to both implementations | Greedy transcript identical to `liquid-audio` | Without fix | With fix | | --------------------------------------------- | ----------: | ----------: | | EN F16 | 191/200 | 200/200 | | JP F32 | 187/200 | 200/200 | | JP F16 | 187/200 | 199/200 | The remaining JP F16 difference is a comma and matches the reference implementation's own bf16 output. Mel relative L2 error versus `liquid-audio`: * EN: 3.2% -> ~2e-6 median * JP: 3.9% -> ~2e-6 median **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b11190/llama-b11190-bin-macos-arm64.tar.gz) - macOS Apple Silicon (arm64, KleidiAI enabled) [DISABLED](https://github.com/ggml-org/llama.cpp/pull/23780) - [macOS Intel (x64)](https://github.com/ggml-org/llama.cpp/releases/download/b11190/llama-b11190-bin-macos-x64.tar.gz) - [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b11190/llam