diff --git a/libavutil/x86/float_dsp.asm b/libavutil/x86/float_dsp.asm
index 517fd63638..817a1bfb52 100644
--- a/libavutil/x86/float_dsp.asm
+++ b/libavutil/x86/float_dsp.asm
@@ -439,29 +439,54 @@ VECTOR_FMUL_REVERSE
 %endif

 ; float scalarproduct_float_sse(const float *v1, const float *v2, int len)
-INIT_XMM sse
-cglobal scalarproduct_float, 3,3,2, v1, v2, offset
+%macro SCALARPRODUCT_FLOAT 0
+cglobal scalarproduct_float, 3,3,3, v1, v2, offset
     shl   offsetd, 2
     add       v1q, offsetq
     add       v2q, offsetq
     neg   offsetq
-    xorps    xmm0, xmm0
+    xorps      m0, m0
 .loop:
-    movaps   xmm1, [v1q+offsetq]
-    mulps    xmm1, [v2q+offsetq]
-    addps    xmm0, xmm1
-    add   offsetq, 16
+    movaps     m1, [v1q+offsetq]
+    mulps      m1, [v2q+offsetq]
+    addps      m0, m1
+    add   offsetq, mmsize
     js .loop
-    movhlps  xmm1, xmm0
-    addps    xmm0, xmm1
-    movss    xmm1, xmm0
-    shufps   xmm0, xmm0, 1
-    addss    xmm0, xmm1
+%if cpuflag(avx)
+    vextractf128 xm2, m0, 0
+    vextractf128 xm3, m0, 1
+    movhlps    xm1, xm2
+    addps      xm2, xm1
+    movss      xm1, xm2
+    shufps     xm2, xm2, 1
+    addss      xm2, xm1
+    movhlps    xm1, xm3
+    addps      xm3, xm1
+    movss      xm1, xm3
+    shufps     xm3, xm3, 1
+    addss      xm3, xm1
+    addss      xm2, xm3
+    movss      xm0, xm2
+%else
+    movhlps    m1, m0
+    addps      m0, m1
+    movss      m1, m0
+    shufps     m0, m0, 1
+    addss      m0, m1
+%endif
 %if ARCH_X86_64 == 0
-    movss     r0m,  xmm0
+    movss     r0m,  m0
     fld dword r0m
 %endif
     RET
+%endmacro
+
+INIT_XMM sse
+SCALARPRODUCT_FLOAT
+%if HAVE_AVX_EXTERNAL
+INIT_YMM avx
+SCALARPRODUCT_FLOAT
+%endif

 ;-----------------------------------------------------------------------------
 ; void ff_butterflies_float(float *src0, float *src1, int len);
diff --git a/libavutil/x86/float_dsp_init.c b/libavutil/x86/float_dsp_init.c
index 8826e4e2c9..abd2472245 100644
--- a/libavutil/x86/float_dsp_init.c
+++ b/libavutil/x86/float_dsp_init.c
@@ -77,6 +77,8 @@ void ff_vector_fmul_reverse_avx2(float *dst, const float *src0,

 float ff_scalarproduct_float_sse(const float *v1, const float *v2, int order);

+float ff_scalarproduct_float_avx(const float *v1, const float *v2, int order);
+
 void ff_butterflies_float_sse(float *av_restrict src0, float *av_restrict src1, int len);

 av_cold void ff_float_dsp_init_x86(AVFloatDSPContext *fdsp)
@@ -109,6 +111,7 @@ av_cold void ff_float_dsp_init_x86(AVFloatDSPContext *fdsp)
         fdsp->vector_dmac_scalar = ff_vector_dmac_scalar_avx;
         fdsp->vector_fmul_add    = ff_vector_fmul_add_avx;
         fdsp->vector_fmul_reverse = ff_vector_fmul_reverse_avx;
+        fdsp->scalarproduct_float = ff_scalarproduct_float_avx;
     }
     if (EXTERNAL_AVX2_FAST(cpu_flags)) {
         fdsp->vector_fmul_reverse = ff_vector_fmul_reverse_avx2;