Skip to content

Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.

benchmark: clarify PyTorch backend (vLLM benchmark coming later)

游雁b34a017fe63d
diff --git a/README.md b/README.md
index 2fa21168c..b7e72f467 100644
--- a/README.md
+++ b/README.md
@@ -34,7 +34,7 @@
<a name="benchmark"></a>
-## Benchmark: Comprehensive ASR Comparison (NVIDIA H100)
+## Benchmark: Comprehensive ASR Comparison (PyTorch, NVIDIA H100)
> 7 models on GPU, 3 models on CPU. 184 long-form Chinese audio files (192 min total). [Full report](https://modelscope.github.io/FunASR/benchmark.html)