whisper with float32 "src0->type == GGML_TYPE_F16 failed"

[mahuichao]

Hello, I am new to whisper.cpp. I convert float32 pytorch model to ggml f32 model,and run the inference.
But I keep getting "src0->type == GGML_TYPE_F16 failed", when backtrace, I found the function "ggml_conv_1d_ph" create GGML_TYPE_F16 tensor in it, and that cause the error, is this a BUG or I miss something important?