Upload 10 files

Melspectrogram_v2.mlmodelc/analytics/coremldata.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b8a8227f0e06c52e639c4335a606c21f0b1a2fe4c1464fdc49dc8adc30abef86
+oid sha256:f9ee8a634d9df8fe10688367fa81db20467e36c4444e82d6bcef70e37e210987
 size 243

Melspectrogram_v2.mlmodelc/coremldata.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6e3b77fcd063d42c86b3d1fa633698de2fb56f2a9a82bd8affba183cba68ea11
-size 610
+oid sha256:aeef2321b385057904ac30af06a109aec6c31513c5189712b14be6f0d026a665
+size 400

Melspectrogram_v2.mlmodelc/metadata.json

@@ -1,14 +1,14 @@
 [
   {
-    "shortDescription" : "Dynamic Mel-Spectrogram Preprocessor (0.1-10s)",
     "metadataOutputVersion" : "3.0",
+    "storagePrecision" : "Float16",
     "outputSchema" : [
       {
         "hasShapeFlexibility" : "0",
         "isOptional" : "0",
         "dataType" : "Float32",
         "formattedType" : "MultiArray (Float32)",
-        "shortDescription" : "128-bin mel-spectrogram features",
+        "shortDescription" : "",
         "shape" : "[]",
         "name" : "melspectrogram",
         "type" : "MultiArray"
@@ -18,19 +18,16 @@
         "isOptional" : "0",
         "dataType" : "Int32",
         "formattedType" : "MultiArray (Int32 1)",
-        "shortDescription" : "Number of valid mel-spectrogram frames",
+        "shortDescription" : "",
         "shape" : "[1]",
         "name" : "melspectrogram_length",
         "type" : "MultiArray"
       }
     ],
-    "version" : "1.0",
     "modelParameters" : [
 
     ],
-    "author" : "FluidAudio",
     "specificationVersion" : 6,
-    "storagePrecision" : "Float16",
     "mlProgramOperationTypeHistogram" : {
       "Range1d" : 2,
       "Gather" : 3,
@@ -60,10 +57,10 @@
       "Mul" : 1
     },
     "computePrecision" : "Mixed (Float16, Float32, Int32)",
+    "isUpdatable" : "0",
     "stateSchema" : [
 
     ],
-    "isUpdatable" : "0",
     "availability" : {
       "macOS" : "12.0",
       "tvOS" : "15.0",
@@ -75,35 +72,35 @@
     "modelType" : {
       "name" : "MLModelType_mlProgram"
     },
+    "userDefinedMetadata" : {
+      "com.github.apple.coremltools.source_dialect" : "TorchScript",
+      "com.github.apple.coremltools.source" : "torch==2.5.0",
+      "com.github.apple.coremltools.version" : "8.3.0"
+    },
     "inputSchema" : [
       {
         "dataType" : "Float32",
         "hasShapeFlexibility" : "1",
         "isOptional" : "0",
-        "shapeFlexibility" : "1 × 1600...160000",
-        "shapeRange" : "[[1, 1], [1600, 160000]]",
-        "formattedType" : "MultiArray (Float32 1 × 1600)",
+        "shapeFlexibility" : "1 × 1...160000",
+        "shapeRange" : "[[1, 1], [1, 160000]]",
+        "formattedType" : "MultiArray (Float32 1 × 1)",
         "type" : "MultiArray",
-        "shape" : "[1, 1600]",
+        "shape" : "[1, 1]",
         "name" : "audio_signal",
-        "shortDescription" : "Raw audio waveform (16kHz, 0.1-10 seconds)"
+        "shortDescription" : ""
       },
       {
         "hasShapeFlexibility" : "0",
         "isOptional" : "0",
         "dataType" : "Int32",
         "formattedType" : "MultiArray (Int32 1)",
-        "shortDescription" : "Number of audio samples",
+        "shortDescription" : "",
         "shape" : "[1]",
         "name" : "audio_length",
         "type" : "MultiArray"
       }
     ],
-    "userDefinedMetadata" : {
-      "com.github.apple.coremltools.source_dialect" : "TorchScript",
-      "com.github.apple.coremltools.source" : "torch==2.5.0",
-      "com.github.apple.coremltools.version" : "8.3.0"
-    },
     "generatedClassName" : "Melspectrogram_v2",
     "method" : "predict"
   }

Melspectrogram_v2.mlmodelc/model.mil

@@ -1,7 +1,7 @@
 program(1.0)
 [buildInfo = dict<tensor<string, []>, tensor<string, []>>({{"coremlc-component-MIL", "3405.2.1"}, {"coremlc-version", "3404.23.1"}, {"coremltools-component-torch", "2.5.0"}, {"coremltools-source-dialect", "TorchScript"}, {"coremltools-version", "8.3.0"}})]
 {
-    func main<ios15>(tensor<int32, [1]> audio_length, tensor<fp32, [1, ?]> audio_signal) [FlexibleShapeInformation = tuple<tuple<tensor<string, []>, dict<tensor<string, []>, tensor<int32, [?]>>>, tuple<tensor<string, []>, dict<tensor<string, []>, list<tensor<int32, [2]>, ?>>>>((("DefaultShapes", {{"audio_signal", [1, 1600]}}), ("RangeDims", {{"audio_signal", [[1, 1], [1600, 160000]]}})))] {
+    func main<ios15>(tensor<int32, [1]> audio_length, tensor<fp32, [1, ?]> audio_signal) [FlexibleShapeInformation = tuple<tuple<tensor<string, []>, dict<tensor<string, []>, tensor<int32, [?]>>>, tuple<tensor<string, []>, dict<tensor<string, []>, list<tensor<int32, [2]>, ?>>>>((("DefaultShapes", {{"audio_signal", [1, 1]}}), ("RangeDims", {{"audio_signal", [[1, 1], [1, 160000]]}})))] {
             tensor<int32, []> var_6 = const()[name = tensor<string, []>("op_6"), val = tensor<int32, []>(512)];
             tensor<int32, [1]> var_7 = add(x = audio_length, y = var_6)[name = tensor<string, []>("op_7")];
             tensor<int32, []> var_9 = const()[name = tensor<string, []>("op_9"), val = tensor<int32, []>(512)];

ParakeetEncoder_v2.mlmodelc/analytics/coremldata.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f833910c665b12464e7a6cbd29f91b559ed928450506f5b28f8422e25619673f
+oid sha256:8cefbb381c134972be702770aac5396402c51d93baf680dcdd5b81189511b1b7
 size 243

ParakeetEncoder_v2.mlmodelc/coremldata.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6c6a83d01a834eae99540b42ee87404447ac18fb1775e8ce4e93fd278304463c
-size 401
+oid sha256:85298a1f0a959c85991a8a83ebcf541de3f84f99173664a5eeb108f16e4c080a
+size 386

ParakeetEncoder_v2.mlmodelc/metadata.json

@@ -0,0 +1,103 @@
+[
+  {
+    "metadataOutputVersion" : "3.0",
+    "storagePrecision" : "Mixed (Float16, Int8)",
+    "outputSchema" : [
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float32",
+        "formattedType" : "MultiArray (Float32 1 × 126 × 1024)",
+        "shortDescription" : "",
+        "shape" : "[1, 126, 1024]",
+        "name" : "encoder_output",
+        "type" : "MultiArray"
+      },
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Int32",
+        "formattedType" : "MultiArray (Int32 1)",
+        "shortDescription" : "",
+        "shape" : "[1]",
+        "name" : "encoder_output_length",
+        "type" : "MultiArray"
+      }
+    ],
+    "modelParameters" : [
+
+    ],
+    "specificationVersion" : 7,
+    "mlProgramOperationTypeHistogram" : {
+      "Transpose" : 171,
+      "Ios16.softmax" : 24,
+      "Split" : 24,
+      "Ios16.linear" : 193,
+      "Ios16.add" : 174,
+      "Select" : 72,
+      "Tile" : 1,
+      "Ios16.sigmoid" : 24,
+      "Ios16.logicalAnd" : 2,
+      "Pad" : 48,
+      "ExpandDims" : 5,
+      "Ios16.constexprAffineDequantize" : 342,
+      "Ios16.silu" : 72,
+      "Ios16.cast" : 4,
+      "Ios16.less" : 1,
+      "Ios16.conv" : 77,
+      "Ios16.layerNorm" : 120,
+      "SliceByIndex" : 48,
+      "Ios16.relu" : 3,
+      "Ios16.matmul" : 72,
+      "Ios16.reshape" : 145,
+      "Ios16.floor" : 3,
+      "Ios16.mul" : 99,
+      "Ios16.logicalNot" : 2
+    },
+    "computePrecision" : "Mixed (Float16, Float32, Int32)",
+    "isUpdatable" : "0",
+    "stateSchema" : [
+
+    ],
+    "availability" : {
+      "macOS" : "13.0",
+      "tvOS" : "16.0",
+      "visionOS" : "1.0",
+      "watchOS" : "9.0",
+      "iOS" : "16.0",
+      "macCatalyst" : "16.0"
+    },
+    "modelType" : {
+      "name" : "MLModelType_mlProgram"
+    },
+    "userDefinedMetadata" : {
+      "com.github.apple.coremltools.source_dialect" : "TorchScript",
+      "com.github.apple.coremltools.source" : "torch==2.5.0",
+      "com.github.apple.coremltools.version" : "8.3.0"
+    },
+    "inputSchema" : [
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float32",
+        "formattedType" : "MultiArray (Float32 1 × 128 × 1001)",
+        "shortDescription" : "",
+        "shape" : "[1, 128, 1001]",
+        "name" : "audio_signal",
+        "type" : "MultiArray"
+      },
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Int32",
+        "formattedType" : "MultiArray (Int32 1)",
+        "shortDescription" : "",
+        "shape" : "[1]",
+        "name" : "length",
+        "type" : "MultiArray"
+      }
+    ],
+    "generatedClassName" : "ParakeetEncoder_v2",
+    "method" : "predict"
+  }
+]

ParakeetEncoder_v2.mlmodelc/weights/weight.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dcc9e458812c178cbdbbf35f7cda2f62b860218ce8aa50bbff811c3e043125b1
-size 591157632
+oid sha256:23bd5658d6e4c716207873b3275fc6d365ce1a027eb8791eaada176fb67abf86
+size 591108480