Add MFB to MIVisisonX

ROCm · Mar 25, 2024 · dc4200b · dc4200b
1 parent dba24ba
commit dc4200b
Show file tree

Hide file tree

Showing 7 changed files with 309 additions and 10 deletions.
diff --git a/amd_openvx_extensions/amd_rpp/CMakeLists.txt b/amd_openvx_extensions/amd_rpp/CMakeLists.txt
@@ -159,6 +159,7 @@ list(APPEND SOURCES
         source/tensor/WarpAffine.cpp
         source/tensor/SequenceRearrange.cpp
         source/tensor/PreemphasisFilter.cpp
+        source/tensor/MelFilterBank.cpp
         source/kernel_rpp.cpp
         source/internal_publishKernels.cpp
     )

diff --git a/amd_openvx_extensions/amd_rpp/include/internal_publishKernels.h b/amd_openvx_extensions/amd_rpp/include/internal_publishKernels.h
@@ -157,6 +157,7 @@ vx_status WarpAffine_Register(vx_context);
 vx_status SequenceRearrange_Register(vx_context);
 vx_status PreemphasisFilter_Register(vx_context);
 vx_status Spectrogram_Register(vx_context);
+vx_status MelFilterBank_Register(vx_context);
 
 // kernel names
 #define VX_KERNEL_RPP_NOPBATCHPD_NAME                           "org.rpp.NopbatchPD"
@@ -283,5 +284,6 @@ vx_status Spectrogram_Register(vx_context);
 #define VX_KERNEL_RPP_SEQUENCEREARRANGE_NAME                    "org.rpp.SequenceRearrange"
 #define VX_KERNEL_RPP_PREEMPHASISFILTER_NAME                    "org.rpp.PreemphasisFilter"
 #define VX_KERNEL_RPP_SPECTROGRAM_NAME                          "org.rpp.Spectrogram"
+#define VX_KERNEL_RPP_MELFILTERBANK_NAME                        "org.rpp.MelFilterBank"
 
 #endif //_AMDVX_EXT__PUBLISH_KERNELS_H_
diff --git a/amd_openvx_extensions/amd_rpp/include/kernels_rpp.h b/amd_openvx_extensions/amd_rpp/include/kernels_rpp.h
@@ -149,7 +149,8 @@ extern "C"
         VX_KERNEL_RPP_VIGNETTE = VX_KERNEL_BASE(VX_ID_AMD, VX_LIBRARY_RPP) + 0x72,
         VX_KERNEL_RPP_WARPAFFINE = VX_KERNEL_BASE(VX_ID_AMD, VX_LIBRARY_RPP) + 0x73,
         VX_KERNEL_RPP_PREEMPHASISFILTER = VX_KERNEL_BASE(VX_ID_AMD, VX_LIBRARY_RPP) + 0x74,
-        VX_KERNEL_RPP_SPECTROGRAM = VX_KERNEL_BASE(VX_ID_AMD, VX_LIBRARY_RPP) + 0x75
+        VX_KERNEL_RPP_SPECTROGRAM = VX_KERNEL_BASE(VX_ID_AMD, VX_LIBRARY_RPP) + 0x75,
+        VX_KERNEL_RPP_MELFILTERBANK = VX_KERNEL_BASE(VX_ID_AMD, VX_LIBRARY_RPP) + 0x76,
     };
 
 #ifdef __cplusplus

diff --git a/amd_openvx_extensions/amd_rpp/include/vx_ext_rpp.h b/amd_openvx_extensions/amd_rpp/include/vx_ext_rpp.h
@@ -1881,23 +1881,41 @@ extern "C"
 	/*! \brief [Graph] Produces a spectrogram from a 1D signal.
 	* \ingroup group_amd_rpp
 	* \param [in] graph The handle to the graph.
-	* \param [in] pSrc The input tensor in <tt>\ref VX_TYPE_UINT8</tt> or <tt>\ref VX_TYPE_FLOAT32</tt> or <tt>\ref VX_TYPE_FLOAT16</tt> or <tt>\ref VX_TYPE_INT8</tt> format data.
-	* \param [in] pSrcLength The input tensor of batch size in <tt>unsigned int<tt> containing the roi values for the input in xywh/ltrb format.
-	* \param [out] pDst The output tensor (begin) in <tt>\ref VX_TYPE_UINT8</tt> or <tt>\ref VX_TYPE_FLOAT32</tt> or <tt>\ref VX_TYPE_FLOAT16</tt> or <tt>\ref VX_TYPE_INT8</tt> format data.
+	* \param [in] pSrc The input tensor in <tt>\ref VX_TYPE_FLOAT32</tt> format data.
+	* \param [in] pSrcRoi The input tensor of batch size in <tt>unsigned int<tt> containing the roi values for the input in xywh (w- samples, h - channels) format.
+	* \param [out] pDst The output tensor (begin) in <tt>\ref VX_TYPE_FLOAT32</tt> format data.
 	* \param [in] pDstDims The input tensor of batch size in <tt>unsigned int<tt> containing the roi values for the output tensor in xywh/ltrb format.
 	* \param [in] windowFn The input array in <tt>\ref VX_TYPE_FLOAT32</tt> format containing the samples of the window function that will be multiplied to each extracted window when calculating the STFT.
-	* \param [in] centerWindow The input scalar in <tt>\ref VX_TYPE_BOOL</tt> format containing indicates whether extracted windows should be padded so that the window function is centered at multiples of window_step.
-	* \param [in] reflectPadding The input scalar in <tt>\ref VX_TYPE_BOOL</tt> format containing indicates the padding policy when sampling outside the bounds of the signal.
+	* \param [in] centerWindow The input scalar in <tt>\ref VX_TYPE_BOOL</tt> format indicating whether extracted windows should be padded so that the window function is centered at multiples of window_step.
+	* \param [in] reflectPadding The input scalar in <tt>\ref VX_TYPE_BOOL</tt> format indicating the padding policy when sampling outside the bounds of the signal.
 	* \param [in] spectrogramLayout The input scalar in <tt>\ref VX_TYPE_INT32</tt> format containing the Output spectrogram layout.
 	* \param [in] power The input scalar in <tt>\ref VX_TYPE_INT32</tt> format containing the exponent of the magnitude of the spectrum.
 	* \param [in] nfft The input scalar in <tt>\ref VX_TYPE_INT32</tt> format containing the size of the FFT.
 	* \param [in] windowLength The input scalar in <tt>\ref VX_TYPE_INT32</tt> format containing Window size in number of samples.
 	* \param [in] windowStep The input array in <tt>\ref VX_TYPE_INT32</tt> format containing the step between the STFT windows in number of samples.
 	* \return A node reference <tt>\ref vx_node</tt>. Any possible errors preventing a successful creation should be checked using <tt>\ref vxGetStatus</tt>.
 	*/
-	SHARED_PUBLIC vx_node VX_API_CALL vxExtRppSpectrogram(vx_graph graph, vx_tensor pSrc, vx_tensor pSrcLength, vx_tensor pDst, vx_tensor pDstDims, vx_array windowFn, vx_scalar centerWindow, vx_scalar reflectPadding, vx_scalar spectrogramLayout,
+	SHARED_PUBLIC vx_node VX_API_CALL vxExtRppSpectrogram(vx_graph graph, vx_tensor pSrc, vx_tensor pSrcRoi, vx_tensor pDst, vx_tensor pDstDims, vx_array windowFn, vx_scalar centerWindow, vx_scalar reflectPadding, vx_scalar spectrogramLayout,
 														  vx_scalar power, vx_scalar nfft, vx_scalar windowLength, vx_scalar windowStep);
 
+	/*! \brief [Graph] Produces a mel-spectrogram from spectrogram on applying a bank of triangular filters
+	* \ingroup group_amd_rpp
+	* \param [in] graph The handle to the graph.
+	* \param [in] pSrc The input tensor in <tt>\ref VX_TYPE_FLOAT32</tt> format data.
+	* \param [in] pSrcRoi The input tensor of batch size in <tt>unsigned int<tt> containing the roi values for the input in xywh (w- samples, h - channels) format.
+	* \param [out] pDst The output tensor (begin) in <tt>\ref VX_TYPE_FLOAT32</tt> format data.
+	* \param [in] pDstRoi The input tensor of batch size in <tt>unsigned int<tt> containing the roi values for the input in xywh (w- samples, h - channels) format.
+	* \param [in] freqHigh The input array in <tt>\ref VX_TYPE_FLOAT32</tt> format containing the maximum frequency.
+	* \param [in] freqLow The input scalar in <tt>\ref VX_TYPE_FLOAT32</tt> format containing the minimum frequency.
+	* \param [in] melFormula The input scalar in <tt>\ref VX_TYPE_INT32</tt> format indicates the formula used to convert frequencies from hertz to mel and vice versa.
+	* \param [in] nfilter The input scalar in <tt>\ref VX_TYPE_INT32</tt> format containing the number of mel filters.
+	* \param [in] normalize The input scalar in <tt>\ref VX_TYPE_BOOL</tt> format to determine weather to normalize the triangular filter weights by the width of the frequecny bands.
+	* \param [in] sampleRate The input scalar in <tt>\ref VX_TYPE_FLOAT32</tt> format containing the sampling rate of the audio data.
+	* \return A node reference <tt>\ref vx_node</tt>. Any possible errors preventing a successful creation should be checked using <tt>\ref vxGetStatus</tt>.
+	*/
+	SHARED_PUBLIC vx_node VX_API_CALL vxExtRppMelFilterBank(vx_graph graph, vx_tensor pSrc, vx_tensor pSrcRoi, vx_tensor pDst, vx_tensor pDstRoi, vx_scalar freqHigh,
+															vx_scalar freqLow, vx_scalar melFormula, vx_scalar nfilter, vx_scalar normalize, vx_scalar sampleRate);
+
 #ifdef __cplusplus
 }
 #endif

diff --git a/amd_openvx_extensions/amd_rpp/source/internal_publishKernels.cpp b/amd_openvx_extensions/amd_rpp/source/internal_publishKernels.cpp
@@ -162,6 +162,7 @@ vx_status get_kernels_to_publish()
     STATUS_ERROR_CHECK(ADD_KERNEL(WarpAffine_Register));
     STATUS_ERROR_CHECK(ADD_KERNEL(PreemphasisFilter_Register));
     STATUS_ERROR_CHECK(ADD_KERNEL(Spectrogram_Register));
+    STATUS_ERROR_CHECK(ADD_KERNEL(MelFilterBank_Register));
 
     return status;
 }

diff --git a/amd_openvx_extensions/amd_rpp/source/kernel_rpp.cpp b/amd_openvx_extensions/amd_rpp/source/kernel_rpp.cpp
@@ -2559,7 +2559,7 @@ VX_API_ENTRY vx_node VX_API_CALL vxExtRppPreemphasisFilter(vx_graph graph, vx_te
     return node;
 }
 
-VX_API_ENTRY vx_node VX_API_CALL vxExtRppSpectrogram(vx_graph graph, vx_tensor pSrc, vx_tensor pSrcLength, vx_tensor pDst, vx_tensor pDstDims, vx_array windowFn, vx_scalar centerWindows, vx_scalar reflectPadding, vx_scalar spectrogramLayout,
+VX_API_ENTRY vx_node VX_API_CALL vxExtRppSpectrogram(vx_graph graph, vx_tensor pSrc, vx_tensor pSrcRoi, vx_tensor pDst, vx_tensor pDstRoi, vx_array windowFn, vx_scalar centerWindows, vx_scalar reflectPadding, vx_scalar spectrogramLayout,
                                                      vx_scalar power, vx_scalar nfft, vx_scalar windowLength, vx_scalar windowStep) {
     vx_node node = NULL;
     vx_context context = vxGetContext((vx_reference)graph);
@@ -2568,9 +2568,9 @@ VX_API_ENTRY vx_node VX_API_CALL vxExtRppSpectrogram(vx_graph graph, vx_tensor p
         vx_scalar deviceType = vxCreateScalar(vxGetContext((vx_reference)graph), VX_TYPE_UINT32, &devtype);
         vx_reference params[] = {
             (vx_reference)pSrc,
-            (vx_reference)pSrcLength,
+            (vx_reference)pSrcRoi,
             (vx_reference)pDst,
-            (vx_reference)pDstDims,
+            (vx_reference)pDstRoi,
             (vx_reference)windowFn,
             (vx_reference)centerWindows,
             (vx_reference)reflectPadding,
@@ -2585,6 +2585,30 @@ VX_API_ENTRY vx_node VX_API_CALL vxExtRppSpectrogram(vx_graph graph, vx_tensor p
     return node;
 }
 
+VX_API_ENTRY vx_node VX_API_CALL vxExtRppMelFilterBank(vx_graph graph, vx_tensor pSrc, vx_tensor pSrcRoi, vx_tensor pDst, vx_tensor pDstRoi, vx_scalar freqHigh, vx_scalar freqLow, vx_scalar melFormula,
+                                                       vx_scalar nfilter, vx_scalar normalize, vx_scalar sampleRate) {
+    vx_node node = NULL;
+    vx_context context = vxGetContext((vx_reference)graph);
+    if (vxGetStatus((vx_reference)context) == VX_SUCCESS) {
+        vx_uint32 devtype = getGraphAffinity(graph);
+        vx_scalar deviceType = vxCreateScalar(vxGetContext((vx_reference)graph), VX_TYPE_UINT32, &devtype);
+        vx_reference params[] = {
+            (vx_reference)pSrc,
+            (vx_reference)pSrcRoi,
+            (vx_reference)pDst,
+            (vx_reference)pDstRoi,
+            (vx_reference)freqHigh,
+            (vx_reference)freqLow,
+            (vx_reference)melFormula,
+            (vx_reference)nfilter,
+            (vx_reference)normalize,
+            (vx_reference)sampleRate,
+            (vx_reference)deviceType};
+        node = createNode(graph, VX_KERNEL_RPP_MELFILTERBANK, params, 11);
+    }
+    return node;
+}
+
 RpptDataType getRpptDataType(vx_enum vxDataType) {
     switch(vxDataType) {
         case vx_type_e::VX_TYPE_FLOAT32: