compressed models

README.md

@@ -12,8 +12,8 @@ tags:
 - asr
 - quantized
 - automatic-speech-recognition
-extra_gated_heading: "Request Access to WhisperKit Pro (Part of Argmax SDK)"
-extra_gated_description: "WhisperKit Pro is the commercial tier of [WhisperKit](https://github.com/argmaxinc/WhisperKit). Please submit your information below or directly send an email to [whisperkitpro@argmaxinc.com](mailto:whisperkitpro@argmaxinc.com)"
+extra_gated_heading: "WhisperKit Pro is now in early access!"
+extra_gated_description: "WhisperKit Pro is the commercial tier of [WhisperKit](https://github.com/argmaxinc/WhisperKit). Please submit your information below to request early access or directly send an email to [earlyaccess@argmaxinc.com](mailto:earlyaccess@argmaxinc.com)"
 extra_gated_fields:
   Company: text
   Work email: text

config.json

@@ -1,17 +1,9 @@
 {
-    "name": "whisperkit-pro",
-    "version": "0.3",
+    "name": "whisperkit-coreml",
+    "version": "0.2",
     "device_support": [
         {
-            "chips": "A12, A13, S9, S10",
-            "identifiers": [
-                "iPhone11",
-                "iPhone12",
-                "iPad12,1",
-                "iPad12,2",
-                "Watch7",
-                "Watch8"
-            ],
+            "identifiers": ["iPhone11", "iPhone12", "Watch7", "Watch8"],
             "models": {
                 "default": "openai_whisper-tiny",
                 "supported": [
@@ -23,14 +15,7 @@
             }
         },
         {
-            "chips": "A14",
-            "identifiers": [
-                "iPhone13",
-                "iPad13,1",
-                "iPad13,2",
-                "iPad13,18",
-                "iPad13,19"
-            ],
+            "identifiers": ["iPhone13", "iPad13,18", "iPad13,1"],
             "models": {
                 "default": "openai_whisper-base",
                 "supported": [
@@ -39,25 +24,18 @@
                     "openai_whisper-base",
                     "openai_whisper-base.en",
                     "openai_whisper-small",
-                    "openai_whisper-small.en",
-                    "openai_whisper-small_216MB",
-                    "openai_whisper-small.en_217MB"
+                    "openai_whisper-small.en"
                 ]
             }
         },
         {
-            "chips": "A15, A16, A17 Pro, A18, A18 Pro",
             "identifiers": [
                 "iPhone14",
                 "iPhone15",
                 "iPhone16",
                 "iPhone17",
                 "iPad14,1",
-                "iPad14,2",
-                "iPad15,7",
-                "iPad15,8",
-                "iPad16,1",
-                "iPad16,2"
+                "iPad14,2"
             ],
             "models": {
                 "default": "openai_whisper-base",
@@ -68,38 +46,23 @@
                     "openai_whisper-base.en",
                     "openai_whisper-small",
                     "openai_whisper-small.en",
-                    "openai_whisper-small_216MB",
-                    "openai_whisper-small.en_217MB",
-                    "openai_whisper-large-v3-v20240930_626MB"
                 ]
             }
         },
         {
-            "chips": "M1",
             "identifiers": [
-                "MacBookPro17,1",
-                "MacBookPro18,1",
-                "MacBookPro18,2",
-                "MacBookPro18,3",
-                "MacBookPro18,4",
-                "MacBookAir10,1",
-                "Macmini9,1",
-                "iMac21,1",
-                "iMac21,2",
                 "Mac13",
-                "iPad13,4",
-                "iPad13,5",
-                "iPad13,6",
-                "iPad13,7",
-                "iPad13,8",
-                "iPad13,9",
-                "iPad13,10",
-                "iPad13,11",
+                "iMac21",
+                "MacBookAir10,1",
+                "MacBookPro17",
+                "MacBookPro18",
+                "Macmini9",
                 "iPad13,16",
-                "iPad13,17"
+                "iPad13,4",
+                "iPad13,8"
             ],
             "models": {
-                "default": "openai_whisper-large-v3-v20240930_626MB",
+                "default": "openai_whisper-large-v3-v20240930",
                 "supported": [
                     "openai_whisper-tiny",
                     "openai_whisper-tiny.en",
@@ -107,16 +70,12 @@
                     "openai_whisper-base.en",
                     "openai_whisper-small",
                     "openai_whisper-small.en",
-                    "openai_whisper-small_216MB",
-                    "openai_whisper-small.en_217MB",
                     "distil-whisper_distil-large-v3",
                     "openai_whisper-large-v3-v20240930",
-                    "openai_whisper-large-v3-v20240930_626MB"
                 ]
             }
         },
         {
-            "chips": "M2, M3, M4",
             "identifiers": [
                 "Mac14",
                 "Mac15",
@@ -129,7 +88,6 @@
                 "iPad14,9",
                 "iPad14,10",
                 "iPad14,11",
-                "iPad15",
                 "iPad16"
             ],
             "models": {
@@ -141,30 +99,14 @@
                     "openai_whisper-base.en",
                     "openai_whisper-small",
                     "openai_whisper-small.en",
-                    "openai_whisper-small_216MB",
-                    "openai_whisper-small.en_217MB",
                     "distil-whisper_distil-large-v3",
                     "distil-whisper_distil-large-v3_turbo",
                     "openai_whisper-large-v3-v20240930",
-                    "openai_whisper-large-v3-v20240930_626MB",
                     "openai_whisper-large-v3-v20240930_turbo",
                 ]
             }
         }
     ],
     "model_checksums": {
-      "openai_whisper-tiny.en": "8fb84af1ddf1111317536b1f76cf7212",
-      "openai_whisper-small_216MB": "2ad52b77916bb98955c476f58d3dad20",
-      "distil-whisper_distil-large-v3": "c87be28423880a465532fe46a9a5c661",
-      "openai_whisper-large-v3-v20240930": "5d295e85eff97f068fb52cee470d7c08",
-      "openai_whisper-tiny": "9568b121b209b755e37ca17902108e3a",
-      "openai_whisper-small": "9e8c4410a7c763f459e43bbb9a19a917",
-      "openai_whisper-small.en_217MB": "c37db641db4c35864a62035ad98ac6ac",
-      "openai_whisper-large-v3-v20240930_turbo": "4fe350b7b8a53385c7038602ef8c4831",
-      "openai_whisper-large-v3-v20240930_626MB": "3809f2ea741bc265e7abed13e38e0546",
-      "openai_whisper-base.en": "59122ac050f5ca2e99908156cb3a3a67",
-      "openai_whisper-small.en": "6ac617d05682bdae030ee3acd833ebcb",
-      "distil-whisper_distil-large-v3_turbo": "b3fc9b66ae7f86426595f803e4c3a466",
-      "openai_whisper-base": "a486d0f75d21035e3863a5766028e90c"
     }
 }

distil-whisper_distil-large-v3/AudioEncoder.mlmodelc/model.mlmodel

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:392b6bd6a2e983ace4d69ab7afd31bfb48520348e61630b7c97aefc13f0b1149
-size 448971

distil-whisper_distil-large-v3/TextDecoder.mlmodelc/model.mlmodel

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:930381b0ad37a3e9d71635c6a9ee481d74fcef7b2343c5bf86fead4b15671e84
-size 82570

distil-whisper_distil-large-v3_turbo/AudioEncoder.mlmodelc/model.mlmodel

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:392b6bd6a2e983ace4d69ab7afd31bfb48520348e61630b7c97aefc13f0b1149
-size 448971

distil-whisper_distil-large-v3_turbo/TextDecoder.mlmodelc/model.mlmodel

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:930381b0ad37a3e9d71635c6a9ee481d74fcef7b2343c5bf86fead4b15671e84
-size 82570

openai_whisper-base.en/AudioEncoder.mlmodelc/model.mlmodel

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:7a20024738e5e013d5375ef535b8ddd4addfca7fca038f9d3aebc2f5920e5a0f
-size 104317

openai_whisper-base.en/TextDecoder.mlmodelc/model.mlmodel

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:3971516f726ebf25d38340550d9d26e25c72ec646db8846dcdb9b18c7e66196a
-size 152960

openai_whisper-base.en/generation_config.json

@@ -1 +1 @@
-{"alignment_heads": [[2, 0], [3, 1], [3, 3], [4, 7]], "begin_suppress_tokens": [220, 50256], "bos_token_id": 50257, "decoder_start_token_id": 50257, "eos_token_id": 50256, "forced_decoder_ids": [[1, 50362]], "is_multilingual": false, "max_initial_timestamp_index": 50, "max_length": 448, "no_timestamps_token_id": 50362, "pad_token_id": 50256, "prev_sot_token_id": 50360, "return_timestamps": false, "suppress_tokens": [1, 2, 7, 8, 9, 10, 14, 25, 26, 27, 28, 29, 31, 58, 59, 60, 61, 62, 63, 90, 91, 92, 93, 357, 366, 438, 532, 685, 705, 796, 930, 1058, 1220, 1267, 1279, 1303, 1343, 1377, 1391, 1635, 1782, 1875, 2162, 2361, 2488, 3467, 4008, 4211, 4600, 4808, 5299, 5855, 6329, 7203, 9609, 9959, 10563, 10786, 11420, 11709, 11907, 13163, 13697, 13700, 14808, 15306, 16410, 16791, 17992, 19203, 19510, 20724, 22305, 22935, 27007, 30109, 30420, 33409, 34949, 40283, 40493, 40549, 47282, 49146, 50257, 50357, 50358, 50359, 50360, 50361], "transformers_version": "4.31.0.dev0"}
+{"alignment_heads": [[3, 3], [4, 7], [5, 1], [5, 5], [5, 7]], "begin_suppress_tokens": [220, 50256], "bos_token_id": 50257, "decoder_start_token_id": 50257, "eos_token_id": 50256, "forced_decoder_ids": [[1, 50362]], "is_multilingual": false, "max_initial_timestamp_index": 50, "max_length": 448, "no_timestamps_token_id": 50362, "pad_token_id": 50256, "prev_sot_token_id": 50360, "return_timestamps": false, "suppress_tokens": [1, 2, 7, 8, 9, 10, 14, 25, 26, 27, 28, 29, 31, 58, 59, 60, 61, 62, 63, 90, 91, 92, 93, 357, 366, 438, 532, 685, 705, 796, 930, 1058, 1220, 1267, 1279, 1303, 1343, 1377, 1391, 1635, 1782, 1875, 2162, 2361, 2488, 3467, 4008, 4211, 4600, 4808, 5299, 5855, 6329, 7203, 9609, 9959, 10563, 10786, 11420, 11709, 11907, 13163, 13697, 13700, 14808, 15306, 16410, 16791, 17992, 19203, 19510, 20724, 22305, 22935, 27007, 30109, 30420, 33409, 34949, 40283, 40493, 40549, 47282, 49146, 50257, 50357, 50358, 50359, 50360, 50361], "transformers_version": "4.31.0.dev0"}

openai_whisper-base/AudioEncoder.mlmodelc/model.mlmodel

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:7a20024738e5e013d5375ef535b8ddd4addfca7fca038f9d3aebc2f5920e5a0f
-size 104317

openai_whisper-base/TextDecoder.mlmodelc/model.mlmodel

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:1eb1cce138ab546a2a707f90df56e44b16b56985bba928b70a9877327b8b2f23
-size 157094

openai_whisper-base/generation_config.json

@@ -1 +1 @@
-{"alignment_heads": [[3, 0], [2, 2], [4, 7], [3, 1], [4, 2], [0, 5], [5, 1], [5, 6]], "begin_suppress_tokens": [220, 50257], "bos_token_id": 50257, "decoder_start_token_id": 50258, "eos_token_id": 50257, "forced_decoder_ids": [[1, null], [2, 50359]], "is_multilingual": true, "lang_to_id": {"<|af|>": 50327, "<|am|>": 50334, "<|ar|>": 50272, "<|as|>": 50350, "<|az|>": 50304, "<|ba|>": 50355, "<|be|>": 50330, "<|bg|>": 50292, "<|bn|>": 50302, "<|bo|>": 50347, "<|br|>": 50309, "<|bs|>": 50315, "<|ca|>": 50270, "<|cs|>": 50283, "<|cy|>": 50297, "<|da|>": 50285, "<|de|>": 50261, "<|el|>": 50281, "<|en|>": 50259, "<|es|>": 50262, "<|et|>": 50307, "<|eu|>": 50310, "<|fa|>": 50300, "<|fi|>": 50277, "<|fo|>": 50338, "<|fr|>": 50265, "<|gl|>": 50319, "<|gu|>": 50333, "<|haw|>": 50352, "<|ha|>": 50354, "<|he|>": 50279, "<|hi|>": 50276, "<|hr|>": 50291, "<|ht|>": 50339, "<|hu|>": 50286, "<|hy|>": 50312, "<|id|>": 50275, "<|is|>": 50311, "<|it|>": 50274, "<|ja|>": 50266, "<|jw|>": 50356, "<|ka|>": 50329, "<|kk|>": 50316, "<|km|>": 50323, "<|kn|>": 50306, "<|ko|>": 50264, "<|la|>": 50294, "<|lb|>": 50345, "<|ln|>": 50353, "<|lo|>": 50336, "<|lt|>": 50293, "<|lv|>": 50301, "<|mg|>": 50349, "<|mi|>": 50295, "<|mk|>": 50308, "<|ml|>": 50296, "<|mn|>": 50314, "<|mr|>": 50320, "<|ms|>": 50282, "<|mt|>": 50343, "<|my|>": 50346, "<|ne|>": 50313, "<|nl|>": 50271, "<|nn|>": 50342, "<|no|>": 50288, "<|oc|>": 50328, "<|pa|>": 50321, "<|pl|>": 50269, "<|ps|>": 50340, "<|pt|>": 50267, "<|ro|>": 50284, "<|ru|>": 50263, "<|sa|>": 50344, "<|sd|>": 50332, "<|si|>": 50322, "<|sk|>": 50298, "<|sl|>": 50305, "<|sn|>": 50324, "<|so|>": 50326, "<|sq|>": 50317, "<|sr|>": 50303, "<|su|>": 50357, "<|sv|>": 50273, "<|sw|>": 50318, "<|ta|>": 50287, "<|te|>": 50299, "<|tg|>": 50331, "<|th|>": 50289, "<|tk|>": 50341, "<|tl|>": 50348, "<|tr|>": 50268, "<|tt|>": 50351, "<|uk|>": 50280, "<|ur|>": 50290, "<|uz|>": 50337, "<|vi|>": 50278, "<|yi|>": 50335, "<|yo|>": 50325, "<|zh|>": 50260}, "max_initial_timestamp_index": 50, "max_length": 448, "no_timestamps_token_id": 50363, "pad_token_id": 50257, "prev_sot_token_id": 50361, "return_timestamps": false, "suppress_tokens": [1, 2, 7, 8, 9, 10, 14, 25, 26, 27, 28, 29, 31, 58, 59, 60, 61, 62, 63, 90, 91, 92, 93, 359, 503, 522, 542, 873, 893, 902, 918, 922, 931, 1350, 1853, 1982, 2460, 2627, 3246, 3253, 3268, 3536, 3846, 3961, 4183, 4667, 6585, 6647, 7273, 9061, 9383, 10428, 10929, 11938, 12033, 12331, 12562, 13793, 14157, 14635, 15265, 15618, 16553, 16604, 18362, 18956, 20075, 21675, 22520, 26130, 26161, 26435, 28279, 29464, 31650, 32302, 32470, 36865, 42863, 47425, 49870, 50254, 50258, 50358, 50359, 50360, 50361, 50362], "task_to_id": {"transcribe": 50359, "translate": 50358}, "transformers_version": "4.31.0.dev0"}
+{"alignment_heads": [[3, 1], [4, 2], [4, 3], [4, 7], [5, 1], [5, 2], [5, 4], [5, 6]], "begin_suppress_tokens": [220, 50257], "bos_token_id": 50257, "decoder_start_token_id": 50258, "eos_token_id": 50257, "forced_decoder_ids": [[1, null], [2, 50359]], "is_multilingual": true, "lang_to_id": {"<|af|>": 50327, "<|am|>": 50334, "<|ar|>": 50272, "<|as|>": 50350, "<|az|>": 50304, "<|ba|>": 50355, "<|be|>": 50330, "<|bg|>": 50292, "<|bn|>": 50302, "<|bo|>": 50347, "<|br|>": 50309, "<|bs|>": 50315, "<|ca|>": 50270, "<|cs|>": 50283, "<|cy|>": 50297, "<|da|>": 50285, "<|de|>": 50261, "<|el|>": 50281, "<|en|>": 50259, "<|es|>": 50262, "<|et|>": 50307, "<|eu|>": 50310, "<|fa|>": 50300, "<|fi|>": 50277, "<|fo|>": 50338, "<|fr|>": 50265, "<|gl|>": 50319, "<|gu|>": 50333, "<|haw|>": 50352, "<|ha|>": 50354, "<|he|>": 50279, "<|hi|>": 50276, "<|hr|>": 50291, "<|ht|>": 50339, "<|hu|>": 50286, "<|hy|>": 50312, "<|id|>": 50275, "<|is|>": 50311, "<|it|>": 50274, "<|ja|>": 50266, "<|jw|>": 50356, "<|ka|>": 50329, "<|kk|>": 50316, "<|km|>": 50323, "<|kn|>": 50306, "<|ko|>": 50264, "<|la|>": 50294, "<|lb|>": 50345, "<|ln|>": 50353, "<|lo|>": 50336, "<|lt|>": 50293, "<|lv|>": 50301, "<|mg|>": 50349, "<|mi|>": 50295, "<|mk|>": 50308, "<|ml|>": 50296, "<|mn|>": 50314, "<|mr|>": 50320, "<|ms|>": 50282, "<|mt|>": 50343, "<|my|>": 50346, "<|ne|>": 50313, "<|nl|>": 50271, "<|nn|>": 50342, "<|no|>": 50288, "<|oc|>": 50328, "<|pa|>": 50321, "<|pl|>": 50269, "<|ps|>": 50340, "<|pt|>": 50267, "<|ro|>": 50284, "<|ru|>": 50263, "<|sa|>": 50344, "<|sd|>": 50332, "<|si|>": 50322, "<|sk|>": 50298, "<|sl|>": 50305, "<|sn|>": 50324, "<|so|>": 50326, "<|sq|>": 50317, "<|sr|>": 50303, "<|su|>": 50357, "<|sv|>": 50273, "<|sw|>": 50318, "<|ta|>": 50287, "<|te|>": 50299, "<|tg|>": 50331, "<|th|>": 50289, "<|tk|>": 50341, "<|tl|>": 50348, "<|tr|>": 50268, "<|tt|>": 50351, "<|uk|>": 50280, "<|ur|>": 50290, "<|uz|>": 50337, "<|vi|>": 50278, "<|yi|>": 50335, "<|yo|>": 50325, "<|zh|>": 50260}, "max_initial_timestamp_index": 50, "max_length": 448, "no_timestamps_token_id": 50363, "pad_token_id": 50257, "prev_sot_token_id": 50361, "return_timestamps": false, "suppress_tokens": [1, 2, 7, 8, 9, 10, 14, 25, 26, 27, 28, 29, 31, 58, 59, 60, 61, 62, 63, 90, 91, 92, 93, 359, 503, 522, 542, 873, 893, 902, 918, 922, 931, 1350, 1853, 1982, 2460, 2627, 3246, 3253, 3268, 3536, 3846, 3961, 4183, 4667, 6585, 6647, 7273, 9061, 9383, 10428, 10929, 11938, 12033, 12331, 12562, 13793, 14157, 14635, 15265, 15618, 16553, 16604, 18362, 18956, 20075, 21675, 22520, 26130, 26161, 26435, 28279, 29464, 31650, 32302, 32470, 36865, 42863, 47425, 49870, 50254, 50258, 50358, 50359, 50360, 50361, 50362], "task_to_id": {"transcribe": 50359, "translate": 50358}, "transformers_version": "4.31.0.dev0"}

openai_whisper-large-v3-v20240930/AudioEncoder.mlmodelc/model.mlmodel

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:10e6e324a74f2d2413d7a60d156629aaa9847b46471a9dfdc658db8718cee497
-size 454525

openai_whisper-large-v3-v20240930/TextDecoder.mlmodelc/model.mlmodel

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:3bb88895c408fc7a34c6ba6ab397e112d97d9295bfc52e98d9837046a61e6c74
-size 108884

openai_whisper-large-v3-v20240930/generation_config.json

@@ -1 +1 @@
-{"alignment_heads": [[2, 11], [1, 1], [3, 6], [2, 4], [2, 8], [3, 15]], "begin_suppress_tokens": [220, 50257], "bos_token_id": 50257, "decoder_start_token_id": 50258, "eos_token_id": 50257, "forced_decoder_ids": [[1, null], [2, 50360]], "is_multilingual": true, "lang_to_id": {"<|af|>": 50327, "<|am|>": 50334, "<|ar|>": 50272, "<|as|>": 50350, "<|az|>": 50304, "<|ba|>": 50355, "<|be|>": 50330, "<|bg|>": 50292, "<|bn|>": 50302, "<|bo|>": 50347, "<|br|>": 50309, "<|bs|>": 50315, "<|ca|>": 50270, "<|cs|>": 50283, "<|cy|>": 50297, "<|da|>": 50285, "<|de|>": 50261, "<|el|>": 50281, "<|en|>": 50259, "<|es|>": 50262, "<|et|>": 50307, "<|eu|>": 50310, "<|fa|>": 50300, "<|fi|>": 50277, "<|fo|>": 50338, "<|fr|>": 50265, "<|gl|>": 50319, "<|gu|>": 50333, "<|haw|>": 50352, "<|ha|>": 50354, "<|he|>": 50279, "<|hi|>": 50276, "<|hr|>": 50291, "<|ht|>": 50339, "<|hu|>": 50286, "<|hy|>": 50312, "<|id|>": 50275, "<|is|>": 50311, "<|it|>": 50274, "<|ja|>": 50266, "<|jw|>": 50356, "<|ka|>": 50329, "<|kk|>": 50316, "<|km|>": 50323, "<|kn|>": 50306, "<|ko|>": 50264, "<|la|>": 50294, "<|lb|>": 50345, "<|ln|>": 50353, "<|lo|>": 50336, "<|lt|>": 50293, "<|lv|>": 50301, "<|mg|>": 50349, "<|mi|>": 50295, "<|mk|>": 50308, "<|ml|>": 50296, "<|mn|>": 50314, "<|mr|>": 50320, "<|ms|>": 50282, "<|mt|>": 50343, "<|my|>": 50346, "<|ne|>": 50313, "<|nl|>": 50271, "<|nn|>": 50342, "<|no|>": 50288, "<|oc|>": 50328, "<|pa|>": 50321, "<|pl|>": 50269, "<|ps|>": 50340, "<|pt|>": 50267, "<|ro|>": 50284, "<|ru|>": 50263, "<|sa|>": 50344, "<|sd|>": 50332, "<|si|>": 50322, "<|sk|>": 50298, "<|sl|>": 50305, "<|sn|>": 50324, "<|so|>": 50326, "<|sq|>": 50317, "<|sr|>": 50303, "<|su|>": 50357, "<|sv|>": 50273, "<|sw|>": 50318, "<|ta|>": 50287, "<|te|>": 50299, "<|tg|>": 50331, "<|th|>": 50289, "<|tk|>": 50341, "<|tl|>": 50348, "<|tr|>": 50268, "<|tt|>": 50351, "<|uk|>": 50280, "<|ur|>": 50290, "<|uz|>": 50337, "<|vi|>": 50278, "<|yi|>": 50335, "<|yo|>": 50325, "<|yue|>": 50358, "<|zh|>": 50260}, "max_initial_timestamp_index": 50, "max_length": 448, "no_timestamps_token_id": 50364, "pad_token_id": 50257, "prev_sot_token_id": 50362, "return_timestamps": false, "suppress_tokens": [1, 2, 7, 8, 9, 10, 14, 25, 26, 27, 28, 29, 31, 58, 59, 60, 61, 62, 63, 90, 91, 92, 93, 359, 503, 522, 542, 873, 893, 902, 918, 922, 931, 1350, 1853, 1982, 2460, 2627, 3246, 3253, 3268, 3536, 3846, 3961, 4183, 4667, 6585, 6647, 7273, 9061, 9383, 10428, 10929, 11938, 12033, 12331, 12562, 13793, 14157, 14635, 15265, 15618, 16553, 16604, 18362, 18956, 20075, 21675, 22520, 26130, 26161, 26435, 28279, 29464, 31650, 32302, 32470, 36865, 42863, 47425, 49870, 50254, 50258, 50359, 50360, 50361, 50362, 50363], "task_to_id": {"transcribe": 50360, "translate": 50359}, "transformers_version": "4.46.0.dev0"}
+{"alignment_heads": [[2, 4], [2, 11], [3, 3], [3, 6], [3, 11], [3, 14]], "begin_suppress_tokens": [220, 50257], "bos_token_id": 50257, "decoder_start_token_id": 50258, "eos_token_id": 50257, "forced_decoder_ids": [[1, null], [2, 50360]], "is_multilingual": true, "lang_to_id": {"<|af|>": 50327, "<|am|>": 50334, "<|ar|>": 50272, "<|as|>": 50350, "<|az|>": 50304, "<|ba|>": 50355, "<|be|>": 50330, "<|bg|>": 50292, "<|bn|>": 50302, "<|bo|>": 50347, "<|br|>": 50309, "<|bs|>": 50315, "<|ca|>": 50270, "<|cs|>": 50283, "<|cy|>": 50297, "<|da|>": 50285, "<|de|>": 50261, "<|el|>": 50281, "<|en|>": 50259, "<|es|>": 50262, "<|et|>": 50307, "<|eu|>": 50310, "<|fa|>": 50300, "<|fi|>": 50277, "<|fo|>": 50338, "<|fr|>": 50265, "<|gl|>": 50319, "<|gu|>": 50333, "<|haw|>": 50352, "<|ha|>": 50354, "<|he|>": 50279, "<|hi|>": 50276, "<|hr|>": 50291, "<|ht|>": 50339, "<|hu|>": 50286, "<|hy|>": 50312, "<|id|>": 50275, "<|is|>": 50311, "<|it|>": 50274, "<|ja|>": 50266, "<|jw|>": 50356, "<|ka|>": 50329, "<|kk|>": 50316, "<|km|>": 50323, "<|kn|>": 50306, "<|ko|>": 50264, "<|la|>": 50294, "<|lb|>": 50345, "<|ln|>": 50353, "<|lo|>": 50336, "<|lt|>": 50293, "<|lv|>": 50301, "<|mg|>": 50349, "<|mi|>": 50295, "<|mk|>": 50308, "<|ml|>": 50296, "<|mn|>": 50314, "<|mr|>": 50320, "<|ms|>": 50282, "<|mt|>": 50343, "<|my|>": 50346, "<|ne|>": 50313, "<|nl|>": 50271, "<|nn|>": 50342, "<|no|>": 50288, "<|oc|>": 50328, "<|pa|>": 50321, "<|pl|>": 50269, "<|ps|>": 50340, "<|pt|>": 50267, "<|ro|>": 50284, "<|ru|>": 50263, "<|sa|>": 50344, "<|sd|>": 50332, "<|si|>": 50322, "<|sk|>": 50298, "<|sl|>": 50305, "<|sn|>": 50324, "<|so|>": 50326, "<|sq|>": 50317, "<|sr|>": 50303, "<|su|>": 50357, "<|sv|>": 50273, "<|sw|>": 50318, "<|ta|>": 50287, "<|te|>": 50299, "<|tg|>": 50331, "<|th|>": 50289, "<|tk|>": 50341, "<|tl|>": 50348, "<|tr|>": 50268, "<|tt|>": 50351, "<|uk|>": 50280, "<|ur|>": 50290, "<|uz|>": 50337, "<|vi|>": 50278, "<|yi|>": 50335, "<|yo|>": 50325, "<|yue|>": 50358, "<|zh|>": 50260}, "max_initial_timestamp_index": 50, "max_length": 448, "no_timestamps_token_id": 50364, "pad_token_id": 50257, "prev_sot_token_id": 50362, "return_timestamps": false, "suppress_tokens": [1, 2, 7, 8, 9, 10, 14, 25, 26, 27, 28, 29, 31, 58, 59, 60, 61, 62, 63, 90, 91, 92, 93, 359, 503, 522, 542, 873, 893, 902, 918, 922, 931, 1350, 1853, 1982, 2460, 2627, 3246, 3253, 3268, 3536, 3846, 3961, 4183, 4667, 6585, 6647, 7273, 9061, 9383, 10428, 10929, 11938, 12033, 12331, 12562, 13793, 14157, 14635, 15265, 15618, 16553, 16604, 18362, 18956, 20075, 21675, 22520, 26130, 26161, 26435, 28279, 29464, 31650, 32302, 32470, 36865, 42863, 47425, 49870, 50254, 50258, 50359, 50360, 50361, 50362, 50363], "task_to_id": {"transcribe": 50360, "translate": 50359}, "transformers_version": "4.46.0.dev0"}

openai_whisper-large-v3-v20240930_626MB/AudioEncoder.mlmodelc/analytics/coremldata.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:5804542da569c631a754ccc129c66669acc58ea46cf1f8b6802147a2d5528bb9
-size 243

openai_whisper-large-v3-v20240930_626MB/AudioEncoder.mlmodelc/coremldata.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:4d8eb697b6f277c3262741c03978dac85c9d74efeb56269e949d7410cb80ab84
-size 434

openai_whisper-large-v3-v20240930_626MB/AudioEncoder.mlmodelc/metadata.json

@@ -1,92 +0,0 @@
-[
-  {
-    "metadataOutputVersion" : "3.0",
-    "storagePrecision" : "Mixed (Float16, Palettized (4 bits), Sparse)",
-    "outputSchema" : [
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 1 × 1280 × 1 × 1500)",
-        "shortDescription" : "",
-        "shape" : "[1, 1280, 1, 1500]",
-        "name" : "encoder_output_embeds",
-        "type" : "MultiArray"
-      },
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 4 × 1280 × 1 × 1536)",
-        "shortDescription" : "",
-        "shape" : "[4, 1280, 1, 1536]",
-        "name" : "encoder_attn_key_cache",
-        "type" : "MultiArray"
-      },
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 4 × 1280 × 1 × 1536)",
-        "shortDescription" : "",
-        "shape" : "[4, 1280, 1, 1536]",
-        "name" : "encoder_attn_value_cache",
-        "type" : "MultiArray"
-      }
-    ],
-    "modelParameters" : [
-
-    ],
-    "specificationVersion" : 9,
-    "mlProgramOperationTypeHistogram" : {
-      "Ios18.constexprLutToDense" : 202,
-      "Ios18.constexprSparseToDense" : 200,
-      "Ios18.conv" : 404,
-      "Ios18.matmul" : 64,
-      "Ios18.batchNorm" : 65,
-      "Pad" : 2,
-      "Ios18.gelu" : 34,
-      "Ios18.concat" : 2,
-      "Ios18.add" : 267,
-      "Ios18.softmax" : 32,
-      "Ios18.layerNorm" : 65,
-      "Ios18.reshape" : 128,
-      "Ios18.mul" : 32
-    },
-    "computePrecision" : "Mixed (Float16, Int32)",
-    "isUpdatable" : "0",
-    "stateSchema" : [
-
-    ],
-    "availability" : {
-      "macOS" : "15.0",
-      "tvOS" : "18.0",
-      "visionOS" : "2.0",
-      "watchOS" : "11.0",
-      "iOS" : "18.0",
-      "macCatalyst" : "18.0"
-    },
-    "modelType" : {
-      "name" : "MLModelType_mlProgram"
-    },
-    "userDefinedMetadata" : {
-      "com.github.apple.coremltools.source_dialect" : "TorchScript",
-      "com.github.apple.coremltools.source" : "torch==2.5.1",
-      "com.github.apple.coremltools.version" : "8.0"
-    },
-    "inputSchema" : [
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 1 × 128 × 1 × 3000)",
-        "shortDescription" : "",
-        "shape" : "[1, 128, 1, 3000]",
-        "name" : "melspectrogram_features",
-        "type" : "MultiArray"
-      }
-    ],
-    "generatedClassName" : "AudioEncoderStateful",
-    "method" : "predict"
-  }
-]

openai_whisper-large-v3-v20240930_626MB/AudioEncoder.mlmodelc/model.mlmodel

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:a9ff86aadad04b5a33905359c151be0c2bca6fca7212b6856376508716072906
-size 798066

openai_whisper-large-v3-v20240930_626MB/AudioEncoder.mlmodelc/weights/weight.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:a1513c45d5252aafd4b7a949d783c5e80786766432017a62227d131958c9a30c
-size 430375168

openai_whisper-large-v3-v20240930_626MB/LICENSE_NOTICE.txt

@@ -1,7 +0,0 @@
-Argmax proprietary and confidential. Under NDA.
-
-Copyright 2024 Argmax, Inc. All rights reserved.
-
-Unauthorized access, copying, use, distribution, and or commercialization of this file, via any medium or means is strictly prohibited.
-
-Please contact Argmax for licensing information at [email protected].

openai_whisper-large-v3-v20240930_626MB/MelSpectrogram.mlmodelc/analytics/coremldata.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:0980462db89a546e1e90888ea38e0a5ddf1f1fec84608802cdbb12f8a5cc7215
-size 243

openai_whisper-large-v3-v20240930_626MB/MelSpectrogram.mlmodelc/coremldata.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:6475c6649047ce609e3fe84b2525843c03342820662404540baf28146c174014
-size 329

openai_whisper-large-v3-v20240930_626MB/MelSpectrogram.mlmodelc/metadata.json

@@ -1,74 +0,0 @@
-[
-  {
-    "metadataOutputVersion" : "3.0",
-    "storagePrecision" : "Float16",
-    "outputSchema" : [
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 1 × 128 × 1 × 3000)",
-        "shortDescription" : "",
-        "shape" : "[1, 128, 1, 3000]",
-        "name" : "melspectrogram_features",
-        "type" : "MultiArray"
-      }
-    ],
-    "modelParameters" : [
-
-    ],
-    "specificationVersion" : 9,
-    "mlProgramOperationTypeHistogram" : {
-      "Ios18.mul" : 2,
-      "Ios18.square" : 2,
-      "Ios18.conv" : 2,
-      "Ios18.matmul" : 1,
-      "Ios18.expandDims" : 4,
-      "Ios18.sub" : 1,
-      "Ios18.log" : 1,
-      "Ios18.add" : 3,
-      "Ios18.sliceByIndex" : 1,
-      "Ios18.maximum" : 1,
-      "Ios18.squeeze" : 2,
-      "Ios18.reshape" : 2,
-      "Ios16.reduceMax" : 1,
-      "Identity" : 1,
-      "Pad" : 1
-    },
-    "computePrecision" : "Mixed (Float16, Float32, Int32)",
-    "isUpdatable" : "0",
-    "stateSchema" : [
-
-    ],
-    "availability" : {
-      "macOS" : "15.0",
-      "tvOS" : "18.0",
-      "visionOS" : "2.0",
-      "watchOS" : "11.0",
-      "iOS" : "18.0",
-      "macCatalyst" : "18.0"
-    },
-    "modelType" : {
-      "name" : "MLModelType_mlProgram"
-    },
-    "userDefinedMetadata" : {
-      "com.github.apple.coremltools.source_dialect" : "TorchScript",
-      "com.github.apple.coremltools.source" : "torch==2.5.1",
-      "com.github.apple.coremltools.version" : "8.0"
-    },
-    "inputSchema" : [
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 480000)",
-        "shortDescription" : "",
-        "shape" : "[480000]",
-        "name" : "audio",
-        "type" : "MultiArray"
-      }
-    ],
-    "generatedClassName" : "MelSpectrogram",
-    "method" : "predict"
-  }
-]

openai_whisper-large-v3-v20240930_626MB/MelSpectrogram.mlmodelc/model.mil

@@ -1,66 +0,0 @@
-program(1.3)
-[buildInfo = dict<string, string>({{"coremlc-component-MIL", "3400.43.1"}, {"coremlc-version", "3400.58.2"}, {"coremltools-component-torch", "2.5.1"}, {"coremltools-source-dialect", "TorchScript"}, {"coremltools-version", "8.0"}})]
-{
-    func main<ios18>(tensor<fp16, [480000]> audio) {
-            tensor<int32, [3]> var_10 = const()[name = string("op_10"), val = tensor<int32, [3]>([1, 1, 480000])];
-            tensor<fp16, [1, 1, 480000]> input_1_cast_fp16 = reshape(shape = var_10, x = audio)[name = string("input_1_cast_fp16")];
-            tensor<int32, [6]> input_3_pad_0 = const()[name = string("input_3_pad_0"), val = tensor<int32, [6]>([0, 0, 0, 0, 200, 200])];
-            string input_3_mode_0 = const()[name = string("input_3_mode_0"), val = string("reflect")];
-            fp16 const_1_to_fp16 = const()[name = string("const_1_to_fp16"), val = fp16(0x0p+0)];
-            tensor<fp16, [1, 1, 480400]> input_3_cast_fp16 = pad(constant_val = const_1_to_fp16, mode = input_3_mode_0, pad = input_3_pad_0, x = input_1_cast_fp16)[name = string("input_3_cast_fp16")];
-            tensor<int32, [1]> var_22 = const()[name = string("op_22"), val = tensor<int32, [1]>([480400])];
-            tensor<fp16, [480400]> input_cast_fp16 = reshape(shape = var_22, x = input_3_cast_fp16)[name = string("input_cast_fp16")];
-            tensor<int32, [1]> expand_dims_0_axes_0 = const()[name = string("expand_dims_0_axes_0"), val = tensor<int32, [1]>([0])];
-            tensor<fp16, [1, 480400]> expand_dims_0_cast_fp16 = expand_dims(axes = expand_dims_0_axes_0, x = input_cast_fp16)[name = string("expand_dims_0_cast_fp16")];
-            tensor<int32, [1]> expand_dims_3 = const()[name = string("expand_dims_3"), val = tensor<int32, [1]>([160])];
-            tensor<int32, [1]> expand_dims_4_axes_0 = const()[name = string("expand_dims_4_axes_0"), val = tensor<int32, [1]>([1])];
-            tensor<fp16, [1, 1, 480400]> expand_dims_4_cast_fp16 = expand_dims(axes = expand_dims_4_axes_0, x = expand_dims_0_cast_fp16)[name = string("expand_dims_4_cast_fp16")];
-            string conv_0_pad_type_0 = const()[name = string("conv_0_pad_type_0"), val = string("valid")];
-            tensor<int32, [2]> conv_0_pad_0 = const()[name = string("conv_0_pad_0"), val = tensor<int32, [2]>([0, 0])];
-            tensor<int32, [1]> conv_0_dilations_0 = const()[name = string("conv_0_dilations_0"), val = tensor<int32, [1]>([1])];
-            int32 conv_0_groups_0 = const()[name = string("conv_0_groups_0"), val = int32(1)];
-            tensor<fp16, [201, 1, 400]> expand_dims_1_to_fp16 = const()[name = string("expand_dims_1_to_fp16"), val = tensor<fp16, [201, 1, 400]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(64)))];
-            tensor<fp16, [1, 201, 3001]> conv_0_cast_fp16 = conv(dilations = conv_0_dilations_0, groups = conv_0_groups_0, pad = conv_0_pad_0, pad_type = conv_0_pad_type_0, strides = expand_dims_3, weight = expand_dims_1_to_fp16, x = expand_dims_4_cast_fp16)[name = string("conv_0_cast_fp16")];
-            string conv_1_pad_type_0 = const()[name = string("conv_1_pad_type_0"), val = string("valid")];
-            tensor<int32, [2]> conv_1_pad_0 = const()[name = string("conv_1_pad_0"), val = tensor<int32, [2]>([0, 0])];
-            tensor<int32, [1]> conv_1_dilations_0 = const()[name = string("conv_1_dilations_0"), val = tensor<int32, [1]>([1])];
-            int32 conv_1_groups_0 = const()[name = string("conv_1_groups_0"), val = int32(1)];
-            tensor<fp16, [201, 1, 400]> expand_dims_2_to_fp16 = const()[name = string("expand_dims_2_to_fp16"), val = tensor<fp16, [201, 1, 400]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(160960)))];
-            tensor<fp16, [1, 201, 3001]> conv_1_cast_fp16 = conv(dilations = conv_1_dilations_0, groups = conv_1_groups_0, pad = conv_1_pad_0, pad_type = conv_1_pad_type_0, strides = expand_dims_3, weight = expand_dims_2_to_fp16, x = expand_dims_4_cast_fp16)[name = string("conv_1_cast_fp16")];
-            tensor<int32, [1]> squeeze_0_axes_0 = const()[name = string("squeeze_0_axes_0"), val = tensor<int32, [1]>([0])];
-            tensor<fp16, [201, 3001]> squeeze_0_cast_fp16 = squeeze(axes = squeeze_0_axes_0, x = conv_0_cast_fp16)[name = string("squeeze_0_cast_fp16")];
-            tensor<int32, [1]> squeeze_1_axes_0 = const()[name = string("squeeze_1_axes_0"), val = tensor<int32, [1]>([0])];
-            tensor<fp16, [201, 3001]> squeeze_1_cast_fp16 = squeeze(axes = squeeze_1_axes_0, x = conv_1_cast_fp16)[name = string("squeeze_1_cast_fp16")];
-            tensor<fp16, [201, 3001]> square_0_cast_fp16 = square(x = squeeze_0_cast_fp16)[name = string("square_0_cast_fp16")];
-            tensor<fp16, [201, 3001]> square_1_cast_fp16 = square(x = squeeze_1_cast_fp16)[name = string("square_1_cast_fp16")];
-            tensor<fp16, [201, 3001]> add_1_cast_fp16 = add(x = square_0_cast_fp16, y = square_1_cast_fp16)[name = string("add_1_cast_fp16")];
-            tensor<fp16, [201, 3001]> magnitudes_1_cast_fp16 = identity(x = add_1_cast_fp16)[name = string("magnitudes_1_cast_fp16")];
-            tensor<int32, [2]> magnitudes_begin_0 = const()[name = string("magnitudes_begin_0"), val = tensor<int32, [2]>([0, 0])];
-            tensor<int32, [2]> magnitudes_end_0 = const()[name = string("magnitudes_end_0"), val = tensor<int32, [2]>([201, 3000])];
-            tensor<bool, [2]> magnitudes_end_mask_0 = const()[name = string("magnitudes_end_mask_0"), val = tensor<bool, [2]>([true, false])];
-            tensor<fp16, [201, 3000]> magnitudes_cast_fp16 = slice_by_index(begin = magnitudes_begin_0, end = magnitudes_end_0, end_mask = magnitudes_end_mask_0, x = magnitudes_1_cast_fp16)[name = string("magnitudes_cast_fp16")];
-            bool mel_spec_1_transpose_x_0 = const()[name = string("mel_spec_1_transpose_x_0"), val = bool(false)];
-            bool mel_spec_1_transpose_y_0 = const()[name = string("mel_spec_1_transpose_y_0"), val = bool(false)];
-            tensor<fp16, [128, 201]> mel_filters_to_fp16 = const()[name = string("mel_filters_to_fp16"), val = tensor<fp16, [128, 201]>(BLOBFILE(path = string("@model_path/weights/weight.bin"), offset = uint64(321856)))];
-            tensor<fp16, [128, 3000]> mel_spec_1_cast_fp16 = matmul(transpose_x = mel_spec_1_transpose_x_0, transpose_y = mel_spec_1_transpose_y_0, x = mel_filters_to_fp16, y = magnitudes_cast_fp16)[name = string("mel_spec_1_cast_fp16")];
-            fp16 var_41_to_fp16 = const()[name = string("op_41_to_fp16"), val = fp16(0x1p-24)];
-            tensor<fp16, [128, 3000]> mel_spec_cast_fp16 = add(x = mel_spec_1_cast_fp16, y = var_41_to_fp16)[name = string("mel_spec_cast_fp16")];
-            fp32 log_0_epsilon_0 = const()[name = string("log_0_epsilon_0"), val = fp32(0x1p-149)];
-            tensor<fp16, [128, 3000]> log_0_cast_fp16 = log(epsilon = log_0_epsilon_0, x = mel_spec_cast_fp16)[name = string("log_0_cast_fp16")];
-            fp16 mul_0_y_0_to_fp16 = const()[name = string("mul_0_y_0_to_fp16"), val = fp16(0x1.bccp-2)];
-            tensor<fp16, [128, 3000]> mul_0_cast_fp16 = mul(x = log_0_cast_fp16, y = mul_0_y_0_to_fp16)[name = string("mul_0_cast_fp16")];
-            bool var_44_keep_dims_0 = const()[name = string("op_44_keep_dims_0"), val = bool(false)];
-            fp16 var_44_cast_fp16 = reduce_max(keep_dims = var_44_keep_dims_0, x = mul_0_cast_fp16)[name = string("op_44_cast_fp16")];
-            fp16 var_46_to_fp16 = const()[name = string("op_46_to_fp16"), val = fp16(0x1p+3)];
-            fp16 var_47_cast_fp16 = sub(x = var_44_cast_fp16, y = var_46_to_fp16)[name = string("op_47_cast_fp16")];
-            tensor<fp16, [128, 3000]> log_spec_3_cast_fp16 = maximum(x = mul_0_cast_fp16, y = var_47_cast_fp16)[name = string("log_spec_3_cast_fp16")];
-            fp16 var_50_to_fp16 = const()[name = string("op_50_to_fp16"), val = fp16(0x1p+2)];
-            tensor<fp16, [128, 3000]> var_51_cast_fp16 = add(x = log_spec_3_cast_fp16, y = var_50_to_fp16)[name = string("op_51_cast_fp16")];
-            fp16 _inversed_log_spec_y_0_to_fp16 = const()[name = string("_inversed_log_spec_y_0_to_fp16"), val = fp16(0x1p-2)];
-            tensor<fp16, [128, 3000]> _inversed_log_spec_cast_fp16 = mul(x = var_51_cast_fp16, y = _inversed_log_spec_y_0_to_fp16)[name = string("_inversed_log_spec_cast_fp16")];
-            tensor<int32, [1]> var_55_axes_0 = const()[name = string("op_55_axes_0"), val = tensor<int32, [1]>([0])];
-            tensor<fp16, [1, 128, 3000]> var_55_cast_fp16 = expand_dims(axes = var_55_axes_0, x = _inversed_log_spec_cast_fp16)[name = string("op_55_cast_fp16")];
-            tensor<int32, [1]> var_62_axes_0 = const()[name = string("op_62_axes_0"), val = tensor<int32, [1]>([2])];
-            tensor<fp16, [1, 128, 1, 3000]> melspectrogram_features = expand_dims(axes = var_62_axes_0, x = var_55_cast_fp16)[name = string("op_62_cast_fp16")];
-        } -> (melspectrogram_features);
-}

openai_whisper-large-v3-v20240930_626MB/MelSpectrogram.mlmodelc/weights/weight.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:009d9fb8f6b589accfa08cebf1c712ef07c3405229ce3cfb3a57ee033c9d8a49
-size 373376

openai_whisper-large-v3-v20240930_626MB/TextDecoder.mlmodelc/analytics/coremldata.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:e77577edea445aa84e0b74bedc71812d8321b3861d62fd1ceee924f6f920f20a
-size 243

openai_whisper-large-v3-v20240930_626MB/TextDecoder.mlmodelc/coremldata.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:c4901c5e4249e42e8f37325412eb6fcf9ca9c5e22660271613675afed77cff8f
-size 754

openai_whisper-large-v3-v20240930_626MB/TextDecoder.mlmodelc/metadata.json

@@ -1,185 +0,0 @@
-[
-  {
-    "metadataOutputVersion" : "3.0",
-    "storagePrecision" : "Mixed (Float16, Palettized (4 bits), Sparse)",
-    "outputSchema" : [
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 1 × 1 × 51866)",
-        "shortDescription" : "",
-        "shape" : "[1, 1, 51866]",
-        "name" : "logits",
-        "type" : "MultiArray"
-      },
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 1 × 5120 × 1 × 1)",
-        "shortDescription" : "",
-        "shape" : "[1, 5120, 1, 1]",
-        "name" : "key_cache_updates",
-        "type" : "MultiArray"
-      },
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 1 × 5120 × 1 × 1)",
-        "shortDescription" : "",
-        "shape" : "[1, 5120, 1, 1]",
-        "name" : "value_cache_updates",
-        "type" : "MultiArray"
-      },
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 1 × 1536)",
-        "shortDescription" : "",
-        "shape" : "[1, 1536]",
-        "name" : "alignment_heads_weights",
-        "type" : "MultiArray"
-      }
-    ],
-    "modelParameters" : [
-
-    ],
-    "specificationVersion" : 9,
-    "mlProgramOperationTypeHistogram" : {
-      "Ios18.expandDims" : 8,
-      "Ios18.softmax" : 8,
-      "Ios18.mul" : 16,
-      "Ios18.matmul" : 16,
-      "Ios18.batchNorm" : 13,
-      "Ios16.reduceMean" : 1,
-      "Split" : 2,
-      "Ios18.readState" : 5,
-      "Ios18.gather" : 3,
-      "Ios18.add" : 62,
-      "Ios18.layerNorm" : 13,
-      "Ios18.reshape" : 32,
-      "Ios18.constexprLutToDense" : 32,
-      "Ios18.constexprSparseToDense" : 33,
-      "Ios18.conv" : 64,
-      "Ios18.gelu" : 4,
-      "Ios18.linear" : 1,
-      "Ios18.cast" : 1,
-      "Ios18.transpose" : 1,
-      "Ios18.concat" : 3,
-      "Ios18.sliceByIndex" : 20,
-      "Ios18.squeeze" : 1
-    },
-    "computePrecision" : "Mixed (Float16, Int32, UInt16)",
-    "isUpdatable" : "0",
-    "stateSchema" : [
-      {
-        "dataType" : "Float16",
-        "isOptional" : "0",
-        "formattedType" : "State (Float16 1 × 1536)",
-        "shortDescription" : "",
-        "shape" : "[1, 1536]",
-        "name" : "encoder_attn_key_padding_mask",
-        "type" : "State"
-      },
-      {
-        "dataType" : "Float16",
-        "isOptional" : "0",
-        "formattedType" : "State (Float16 4 × 1280 × 1 × 1536)",
-        "shortDescription" : "",
-        "shape" : "[4, 1280, 1, 1536]",
-        "name" : "encoder_attn_key_cache",
-        "type" : "State"
-      },
-      {
-        "dataType" : "Float16",
-        "isOptional" : "0",
-        "formattedType" : "State (Float16 4 × 1280 × 1 × 1536)",
-        "shortDescription" : "",
-        "shape" : "[4, 1280, 1, 1536]",
-        "name" : "encoder_attn_value_cache",
-        "type" : "State"
-      },
-      {
-        "dataType" : "Float16",
-        "isOptional" : "0",
-        "formattedType" : "State (Float16 4 × 1280 × 1 × 448)",
-        "shortDescription" : "",
-        "shape" : "[4, 1280, 1, 448]",
-        "name" : "self_attn_key_cache",
-        "type" : "State"
-      },
-      {
-        "dataType" : "Float16",
-        "isOptional" : "0",
-        "formattedType" : "State (Float16 4 × 1280 × 1 × 448)",
-        "shortDescription" : "",
-        "shape" : "[4, 1280, 1, 448]",
-        "name" : "self_attn_value_cache",
-        "type" : "State"
-      }
-    ],
-    "availability" : {
-      "macOS" : "15.0",
-      "tvOS" : "18.0",
-      "visionOS" : "2.0",
-      "watchOS" : "11.0",
-      "iOS" : "18.0",
-      "macCatalyst" : "18.0"
-    },
-    "modelType" : {
-      "name" : "MLModelType_mlProgram"
-    },
-    "userDefinedMetadata" : {
-      "com.github.apple.coremltools.source_dialect" : "TorchScript",
-      "com.github.apple.coremltools.source" : "torch==2.5.1",
-      "com.github.apple.coremltools.version" : "8.0"
-    },
-    "inputSchema" : [
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Int32",
-        "formattedType" : "MultiArray (Int32 1)",
-        "shortDescription" : "",
-        "shape" : "[1]",
-        "name" : "input_ids",
-        "type" : "MultiArray"
-      },
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Int32",
-        "formattedType" : "MultiArray (Int32 1)",
-        "shortDescription" : "",
-        "shape" : "[1]",
-        "name" : "cache_length",
-        "type" : "MultiArray"
-      },
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 1 × 448)",
-        "shortDescription" : "",
-        "shape" : "[1, 448]",
-        "name" : "kv_cache_update_mask",
-        "type" : "MultiArray"
-      },
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 1 × 448)",
-        "shortDescription" : "",
-        "shape" : "[1, 448]",
-        "name" : "decoder_key_padding_mask",
-        "type" : "MultiArray"
-      }
-    ],
-    "generatedClassName" : "TextDecoderStateful",
-    "method" : "predict"
-  }
-]

openai_whisper-large-v3-v20240930_626MB/TextDecoder.mlmodelc/model.mlmodel

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:beb44115c7f126e0a9ea1733afa4b3a50609d361c43b14853b89a2da8000d220
-size 163308

openai_whisper-large-v3-v20240930_626MB/TextDecoder.mlmodelc/weights/weight.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:43abea4ea336b9e458398af192796b78180565f3711530c0d8d8d7f192e199ff
-size 193154996

openai_whisper-large-v3-v20240930_626MB/config.json

@@ -1 +0,0 @@
-{"_name_or_path": "/raid/yoach/tmp_whisper_turbo", "activation_dropout": 0.0, "activation_function": "gelu", "apply_spec_augment": false, "architectures": ["WhisperForConditionalGeneration"], "attention_dropout": 0.0, "begin_suppress_tokens": [220, 50256], "bos_token_id": 50257, "classifier_proj_size": 256, "d_model": 1280, "decoder_attention_heads": 20, "decoder_ffn_dim": 5120, "decoder_layerdrop": 0.0, "decoder_layers": 4, "decoder_start_token_id": 50258, "dropout": 0.0, "encoder_attention_heads": 20, "encoder_ffn_dim": 5120, "encoder_layerdrop": 0.0, "encoder_layers": 32, "eos_token_id": 50257, "init_std": 0.02, "is_encoder_decoder": true, "mask_feature_length": 10, "mask_feature_min_masks": 0, "mask_feature_prob": 0.0, "mask_time_length": 10, "mask_time_min_masks": 2, "mask_time_prob": 0.05, "max_source_positions": 1500, "max_target_positions": 448, "median_filter_width": 7, "model_type": "whisper", "num_hidden_layers": 32, "num_mel_bins": 128, "pad_token_id": 50257, "scale_embedding": false, "torch_dtype": "float16", "transformers_version": "4.46.0.dev0", "use_cache": true, "use_weighted_layer_sum": false, "vocab_size": 51866}

openai_whisper-large-v3-v20240930_626MB/generation_config.json

@@ -1 +0,0 @@
-{"alignment_heads": [[2, 11], [1, 1], [3, 6], [2, 4], [2, 8], [3, 15]], "begin_suppress_tokens": [220, 50257], "bos_token_id": 50257, "decoder_start_token_id": 50258, "eos_token_id": 50257, "forced_decoder_ids": [[1, null], [2, 50360]], "is_multilingual": true, "lang_to_id": {"<|af|>": 50327, "<|am|>": 50334, "<|ar|>": 50272, "<|as|>": 50350, "<|az|>": 50304, "<|ba|>": 50355, "<|be|>": 50330, "<|bg|>": 50292, "<|bn|>": 50302, "<|bo|>": 50347, "<|br|>": 50309, "<|bs|>": 50315, "<|ca|>": 50270, "<|cs|>": 50283, "<|cy|>": 50297, "<|da|>": 50285, "<|de|>": 50261, "<|el|>": 50281, "<|en|>": 50259, "<|es|>": 50262, "<|et|>": 50307, "<|eu|>": 50310, "<|fa|>": 50300, "<|fi|>": 50277, "<|fo|>": 50338, "<|fr|>": 50265, "<|gl|>": 50319, "<|gu|>": 50333, "<|haw|>": 50352, "<|ha|>": 50354, "<|he|>": 50279, "<|hi|>": 50276, "<|hr|>": 50291, "<|ht|>": 50339, "<|hu|>": 50286, "<|hy|>": 50312, "<|id|>": 50275, "<|is|>": 50311, "<|it|>": 50274, "<|ja|>": 50266, "<|jw|>": 50356, "<|ka|>": 50329, "<|kk|>": 50316, "<|km|>": 50323, "<|kn|>": 50306, "<|ko|>": 50264, "<|la|>": 50294, "<|lb|>": 50345, "<|ln|>": 50353, "<|lo|>": 50336, "<|lt|>": 50293, "<|lv|>": 50301, "<|mg|>": 50349, "<|mi|>": 50295, "<|mk|>": 50308, "<|ml|>": 50296, "<|mn|>": 50314, "<|mr|>": 50320, "<|ms|>": 50282, "<|mt|>": 50343, "<|my|>": 50346, "<|ne|>": 50313, "<|nl|>": 50271, "<|nn|>": 50342, "<|no|>": 50288, "<|oc|>": 50328, "<|pa|>": 50321, "<|pl|>": 50269, "<|ps|>": 50340, "<|pt|>": 50267, "<|ro|>": 50284, "<|ru|>": 50263, "<|sa|>": 50344, "<|sd|>": 50332, "<|si|>": 50322, "<|sk|>": 50298, "<|sl|>": 50305, "<|sn|>": 50324, "<|so|>": 50326, "<|sq|>": 50317, "<|sr|>": 50303, "<|su|>": 50357, "<|sv|>": 50273, "<|sw|>": 50318, "<|ta|>": 50287, "<|te|>": 50299, "<|tg|>": 50331, "<|th|>": 50289, "<|tk|>": 50341, "<|tl|>": 50348, "<|tr|>": 50268, "<|tt|>": 50351, "<|uk|>": 50280, "<|ur|>": 50290, "<|uz|>": 50337, "<|vi|>": 50278, "<|yi|>": 50335, "<|yo|>": 50325, "<|yue|>": 50358, "<|zh|>": 50260}, "max_initial_timestamp_index": 50, "max_length": 448, "no_timestamps_token_id": 50364, "pad_token_id": 50257, "prev_sot_token_id": 50362, "return_timestamps": false, "suppress_tokens": [1, 2, 7, 8, 9, 10, 14, 25, 26, 27, 28, 29, 31, 58, 59, 60, 61, 62, 63, 90, 91, 92, 93, 359, 503, 522, 542, 873, 893, 902, 918, 922, 931, 1350, 1853, 1982, 2460, 2627, 3246, 3253, 3268, 3536, 3846, 3961, 4183, 4667, 6585, 6647, 7273, 9061, 9383, 10428, 10929, 11938, 12033, 12331, 12562, 13793, 14157, 14635, 15265, 15618, 16553, 16604, 18362, 18956, 20075, 21675, 22520, 26130, 26161, 26435, 28279, 29464, 31650, 32302, 32470, 36865, 42863, 47425, 49870, 50254, 50258, 50359, 50360, 50361, 50362, 50363], "task_to_id": {"transcribe": 50360, "translate": 50359}, "transformers_version": "4.46.0.dev0"}

openai_whisper-large-v3-v20240930_turbo/AudioEncoder.mlmodelc/model.mlmodel

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:10e6e324a74f2d2413d7a60d156629aaa9847b46471a9dfdc658db8718cee497
-size 454525

openai_whisper-large-v3-v20240930_turbo/TextDecoder.mlmodelc/model.mlmodel

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:3bb88895c408fc7a34c6ba6ab397e112d97d9295bfc52e98d9837046a61e6c74
-size 108884

openai_whisper-large-v3-v20240930_turbo/generation_config.json

@@ -1 +1 @@
-{"alignment_heads": [[2, 11], [1, 1], [3, 6], [2, 4], [2, 8], [3, 15]], "begin_suppress_tokens": [220, 50257], "bos_token_id": 50257, "decoder_start_token_id": 50258, "eos_token_id": 50257, "forced_decoder_ids": [[1, null], [2, 50360]], "is_multilingual": true, "lang_to_id": {"<|af|>": 50327, "<|am|>": 50334, "<|ar|>": 50272, "<|as|>": 50350, "<|az|>": 50304, "<|ba|>": 50355, "<|be|>": 50330, "<|bg|>": 50292, "<|bn|>": 50302, "<|bo|>": 50347, "<|br|>": 50309, "<|bs|>": 50315, "<|ca|>": 50270, "<|cs|>": 50283, "<|cy|>": 50297, "<|da|>": 50285, "<|de|>": 50261, "<|el|>": 50281, "<|en|>": 50259, "<|es|>": 50262, "<|et|>": 50307, "<|eu|>": 50310, "<|fa|>": 50300, "<|fi|>": 50277, "<|fo|>": 50338, "<|fr|>": 50265, "<|gl|>": 50319, "<|gu|>": 50333, "<|haw|>": 50352, "<|ha|>": 50354, "<|he|>": 50279, "<|hi|>": 50276, "<|hr|>": 50291, "<|ht|>": 50339, "<|hu|>": 50286, "<|hy|>": 50312, "<|id|>": 50275, "<|is|>": 50311, "<|it|>": 50274, "<|ja|>": 50266, "<|jw|>": 50356, "<|ka|>": 50329, "<|kk|>": 50316, "<|km|>": 50323, "<|kn|>": 50306, "<|ko|>": 50264, "<|la|>": 50294, "<|lb|>": 50345, "<|ln|>": 50353, "<|lo|>": 50336, "<|lt|>": 50293, "<|lv|>": 50301, "<|mg|>": 50349, "<|mi|>": 50295, "<|mk|>": 50308, "<|ml|>": 50296, "<|mn|>": 50314, "<|mr|>": 50320, "<|ms|>": 50282, "<|mt|>": 50343, "<|my|>": 50346, "<|ne|>": 50313, "<|nl|>": 50271, "<|nn|>": 50342, "<|no|>": 50288, "<|oc|>": 50328, "<|pa|>": 50321, "<|pl|>": 50269, "<|ps|>": 50340, "<|pt|>": 50267, "<|ro|>": 50284, "<|ru|>": 50263, "<|sa|>": 50344, "<|sd|>": 50332, "<|si|>": 50322, "<|sk|>": 50298, "<|sl|>": 50305, "<|sn|>": 50324, "<|so|>": 50326, "<|sq|>": 50317, "<|sr|>": 50303, "<|su|>": 50357, "<|sv|>": 50273, "<|sw|>": 50318, "<|ta|>": 50287, "<|te|>": 50299, "<|tg|>": 50331, "<|th|>": 50289, "<|tk|>": 50341, "<|tl|>": 50348, "<|tr|>": 50268, "<|tt|>": 50351, "<|uk|>": 50280, "<|ur|>": 50290, "<|uz|>": 50337, "<|vi|>": 50278, "<|yi|>": 50335, "<|yo|>": 50325, "<|yue|>": 50358, "<|zh|>": 50260}, "max_initial_timestamp_index": 50, "max_length": 448, "no_timestamps_token_id": 50364, "pad_token_id": 50257, "prev_sot_token_id": 50362, "return_timestamps": false, "suppress_tokens": [1, 2, 7, 8, 9, 10, 14, 25, 26, 27, 28, 29, 31, 58, 59, 60, 61, 62, 63, 90, 91, 92, 93, 359, 503, 522, 542, 873, 893, 902, 918, 922, 931, 1350, 1853, 1982, 2460, 2627, 3246, 3253, 3268, 3536, 3846, 3961, 4183, 4667, 6585, 6647, 7273, 9061, 9383, 10428, 10929, 11938, 12033, 12331, 12562, 13793, 14157, 14635, 15265, 15618, 16553, 16604, 18362, 18956, 20075, 21675, 22520, 26130, 26161, 26435, 28279, 29464, 31650, 32302, 32470, 36865, 42863, 47425, 49870, 50254, 50258, 50359, 50360, 50361, 50362, 50363], "task_to_id": {"transcribe": 50360, "translate": 50359}, "transformers_version": "4.46.0.dev0"}
+{"alignment_heads": [[2, 4], [2, 11], [3, 3], [3, 6], [3, 11], [3, 14]], "begin_suppress_tokens": [220, 50257], "bos_token_id": 50257, "decoder_start_token_id": 50258, "eos_token_id": 50257, "forced_decoder_ids": [[1, null], [2, 50360]], "is_multilingual": true, "lang_to_id": {"<|af|>": 50327, "<|am|>": 50334, "<|ar|>": 50272, "<|as|>": 50350, "<|az|>": 50304, "<|ba|>": 50355, "<|be|>": 50330, "<|bg|>": 50292, "<|bn|>": 50302, "<|bo|>": 50347, "<|br|>": 50309, "<|bs|>": 50315, "<|ca|>": 50270, "<|cs|>": 50283, "<|cy|>": 50297, "<|da|>": 50285, "<|de|>": 50261, "<|el|>": 50281, "<|en|>": 50259, "<|es|>": 50262, "<|et|>": 50307, "<|eu|>": 50310, "<|fa|>": 50300, "<|fi|>": 50277, "<|fo|>": 50338, "<|fr|>": 50265, "<|gl|>": 50319, "<|gu|>": 50333, "<|haw|>": 50352, "<|ha|>": 50354, "<|he|>": 50279, "<|hi|>": 50276, "<|hr|>": 50291, "<|ht|>": 50339, "<|hu|>": 50286, "<|hy|>": 50312, "<|id|>": 50275, "<|is|>": 50311, "<|it|>": 50274, "<|ja|>": 50266, "<|jw|>": 50356, "<|ka|>": 50329, "<|kk|>": 50316, "<|km|>": 50323, "<|kn|>": 50306, "<|ko|>": 50264, "<|la|>": 50294, "<|lb|>": 50345, "<|ln|>": 50353, "<|lo|>": 50336, "<|lt|>": 50293, "<|lv|>": 50301, "<|mg|>": 50349, "<|mi|>": 50295, "<|mk|>": 50308, "<|ml|>": 50296, "<|mn|>": 50314, "<|mr|>": 50320, "<|ms|>": 50282, "<|mt|>": 50343, "<|my|>": 50346, "<|ne|>": 50313, "<|nl|>": 50271, "<|nn|>": 50342, "<|no|>": 50288, "<|oc|>": 50328, "<|pa|>": 50321, "<|pl|>": 50269, "<|ps|>": 50340, "<|pt|>": 50267, "<|ro|>": 50284, "<|ru|>": 50263, "<|sa|>": 50344, "<|sd|>": 50332, "<|si|>": 50322, "<|sk|>": 50298, "<|sl|>": 50305, "<|sn|>": 50324, "<|so|>": 50326, "<|sq|>": 50317, "<|sr|>": 50303, "<|su|>": 50357, "<|sv|>": 50273, "<|sw|>": 50318, "<|ta|>": 50287, "<|te|>": 50299, "<|tg|>": 50331, "<|th|>": 50289, "<|tk|>": 50341, "<|tl|>": 50348, "<|tr|>": 50268, "<|tt|>": 50351, "<|uk|>": 50280, "<|ur|>": 50290, "<|uz|>": 50337, "<|vi|>": 50278, "<|yi|>": 50335, "<|yo|>": 50325, "<|yue|>": 50358, "<|zh|>": 50260}, "max_initial_timestamp_index": 50, "max_length": 448, "no_timestamps_token_id": 50364, "pad_token_id": 50257, "prev_sot_token_id": 50362, "return_timestamps": false, "suppress_tokens": [1, 2, 7, 8, 9, 10, 14, 25, 26, 27, 28, 29, 31, 58, 59, 60, 61, 62, 63, 90, 91, 92, 93, 359, 503, 522, 542, 873, 893, 902, 918, 922, 931, 1350, 1853, 1982, 2460, 2627, 3246, 3253, 3268, 3536, 3846, 3961, 4183, 4667, 6585, 6647, 7273, 9061, 9383, 10428, 10929, 11938, 12033, 12331, 12562, 13793, 14157, 14635, 15265, 15618, 16553, 16604, 18362, 18956, 20075, 21675, 22520, 26130, 26161, 26435, 28279, 29464, 31650, 32302, 32470, 36865, 42863, 47425, 49870, 50254, 50258, 50359, 50360, 50361, 50362, 50363], "task_to_id": {"transcribe": 50360, "translate": 50359}, "transformers_version": "4.46.0.dev0"}

openai_whisper-small.en/AudioEncoder.mlmodelc/model.mlmodel

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:821ebe68a594584d7b1c3ed6cbb895642b5d3593952e1d9e89a859191ebe0b39
-size 202745

openai_whisper-small.en/TextDecoder.mlmodelc/model.mlmodel

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:5c9356d78b41cda632e5d4c5f2ba395e948c3c07aa876ca467a883579bdaf334
-size 309252

openai_whisper-small.en/generation_config.json

@@ -1 +1 @@
-{"alignment_heads": [[4, 1], [3, 7], [1, 2], [7, 5], [9, 4], [7, 4]], "begin_suppress_tokens": [220, 50257], "bos_token_id": 50257, "decoder_start_token_id": 50258, "eos_token_id": 50257, "forced_decoder_ids": [[1, null], [2, 50359]], "is_multilingual": true, "lang_to_id": {"<|af|>": 50327, "<|am|>": 50334, "<|ar|>": 50272, "<|as|>": 50350, "<|az|>": 50304, "<|ba|>": 50355, "<|be|>": 50330, "<|bg|>": 50292, "<|bn|>": 50302, "<|bo|>": 50347, "<|br|>": 50309, "<|bs|>": 50315, "<|ca|>": 50270, "<|cs|>": 50283, "<|cy|>": 50297, "<|da|>": 50285, "<|de|>": 50261, "<|el|>": 50281, "<|en|>": 50259, "<|es|>": 50262, "<|et|>": 50307, "<|eu|>": 50310, "<|fa|>": 50300, "<|fi|>": 50277, "<|fo|>": 50338, "<|fr|>": 50265, "<|gl|>": 50319, "<|gu|>": 50333, "<|haw|>": 50352, "<|ha|>": 50354, "<|he|>": 50279, "<|hi|>": 50276, "<|hr|>": 50291, "<|ht|>": 50339, "<|hu|>": 50286, "<|hy|>": 50312, "<|id|>": 50275, "<|is|>": 50311, "<|it|>": 50274, "<|ja|>": 50266, "<|jw|>": 50356, "<|ka|>": 50329, "<|kk|>": 50316, "<|km|>": 50323, "<|kn|>": 50306, "<|ko|>": 50264, "<|la|>": 50294, "<|lb|>": 50345, "<|ln|>": 50353, "<|lo|>": 50336, "<|lt|>": 50293, "<|lv|>": 50301, "<|mg|>": 50349, "<|mi|>": 50295, "<|mk|>": 50308, "<|ml|>": 50296, "<|mn|>": 50314, "<|mr|>": 50320, "<|ms|>": 50282, "<|mt|>": 50343, "<|my|>": 50346, "<|ne|>": 50313, "<|nl|>": 50271, "<|nn|>": 50342, "<|no|>": 50288, "<|oc|>": 50328, "<|pa|>": 50321, "<|pl|>": 50269, "<|ps|>": 50340, "<|pt|>": 50267, "<|ro|>": 50284, "<|ru|>": 50263, "<|sa|>": 50344, "<|sd|>": 50332, "<|si|>": 50322, "<|sk|>": 50298, "<|sl|>": 50305, "<|sn|>": 50324, "<|so|>": 50326, "<|sq|>": 50317, "<|sr|>": 50303, "<|su|>": 50357, "<|sv|>": 50273, "<|sw|>": 50318, "<|ta|>": 50287, "<|te|>": 50299, "<|tg|>": 50331, "<|th|>": 50289, "<|tk|>": 50341, "<|tl|>": 50348, "<|tr|>": 50268, "<|tt|>": 50351, "<|uk|>": 50280, "<|ur|>": 50290, "<|uz|>": 50337, "<|vi|>": 50278, "<|yi|>": 50335, "<|yo|>": 50325, "<|zh|>": 50260}, "max_initial_timestamp_index": 50, "max_length": 448, "no_timestamps_token_id": 50363, "pad_token_id": 50257, "prev_sot_token_id": 50361, "return_timestamps": false, "suppress_tokens": [1, 2, 7, 8, 9, 10, 14, 25, 26, 27, 28, 29, 31, 58, 59, 60, 61, 62, 63, 90, 91, 92, 93, 359, 503, 522, 542, 873, 893, 902, 918, 922, 931, 1350, 1853, 1982, 2460, 2627, 3246, 3253, 3268, 3536, 3846, 3961, 4183, 4667, 6585, 6647, 7273, 9061, 9383, 10428, 10929, 11938, 12033, 12331, 12562, 13793, 14157, 14635, 15265, 15618, 16553, 16604, 18362, 18956, 20075, 21675, 22520, 26130, 26161, 26435, 28279, 29464, 31650, 32302, 32470, 36865, 42863, 47425, 49870, 50254, 50258, 50358, 50359, 50360, 50361, 50362], "task_to_id": {"transcribe": 50359, "translate": 50358}, "transformers_version": "4.31.0.dev0"}
+{"alignment_heads": [[5, 3], [5, 9], [8, 0], [8, 4], [8, 7], [8, 8], [9, 0], [9, 7], [9, 9], [10, 5]], "begin_suppress_tokens": [220, 50257], "bos_token_id": 50257, "decoder_start_token_id": 50258, "eos_token_id": 50257, "forced_decoder_ids": [[1, null], [2, 50359]], "is_multilingual": true, "lang_to_id": {"<|af|>": 50327, "<|am|>": 50334, "<|ar|>": 50272, "<|as|>": 50350, "<|az|>": 50304, "<|ba|>": 50355, "<|be|>": 50330, "<|bg|>": 50292, "<|bn|>": 50302, "<|bo|>": 50347, "<|br|>": 50309, "<|bs|>": 50315, "<|ca|>": 50270, "<|cs|>": 50283, "<|cy|>": 50297, "<|da|>": 50285, "<|de|>": 50261, "<|el|>": 50281, "<|en|>": 50259, "<|es|>": 50262, "<|et|>": 50307, "<|eu|>": 50310, "<|fa|>": 50300, "<|fi|>": 50277, "<|fo|>": 50338, "<|fr|>": 50265, "<|gl|>": 50319, "<|gu|>": 50333, "<|haw|>": 50352, "<|ha|>": 50354, "<|he|>": 50279, "<|hi|>": 50276, "<|hr|>": 50291, "<|ht|>": 50339, "<|hu|>": 50286, "<|hy|>": 50312, "<|id|>": 50275, "<|is|>": 50311, "<|it|>": 50274, "<|ja|>": 50266, "<|jw|>": 50356, "<|ka|>": 50329, "<|kk|>": 50316, "<|km|>": 50323, "<|kn|>": 50306, "<|ko|>": 50264, "<|la|>": 50294, "<|lb|>": 50345, "<|ln|>": 50353, "<|lo|>": 50336, "<|lt|>": 50293, "<|lv|>": 50301, "<|mg|>": 50349, "<|mi|>": 50295, "<|mk|>": 50308, "<|ml|>": 50296, "<|mn|>": 50314, "<|mr|>": 50320, "<|ms|>": 50282, "<|mt|>": 50343, "<|my|>": 50346, "<|ne|>": 50313, "<|nl|>": 50271, "<|nn|>": 50342, "<|no|>": 50288, "<|oc|>": 50328, "<|pa|>": 50321, "<|pl|>": 50269, "<|ps|>": 50340, "<|pt|>": 50267, "<|ro|>": 50284, "<|ru|>": 50263, "<|sa|>": 50344, "<|sd|>": 50332, "<|si|>": 50322, "<|sk|>": 50298, "<|sl|>": 50305, "<|sn|>": 50324, "<|so|>": 50326, "<|sq|>": 50317, "<|sr|>": 50303, "<|su|>": 50357, "<|sv|>": 50273, "<|sw|>": 50318, "<|ta|>": 50287, "<|te|>": 50299, "<|tg|>": 50331, "<|th|>": 50289, "<|tk|>": 50341, "<|tl|>": 50348, "<|tr|>": 50268, "<|tt|>": 50351, "<|uk|>": 50280, "<|ur|>": 50290, "<|uz|>": 50337, "<|vi|>": 50278, "<|yi|>": 50335, "<|yo|>": 50325, "<|zh|>": 50260}, "max_initial_timestamp_index": 50, "max_length": 448, "no_timestamps_token_id": 50363, "pad_token_id": 50257, "prev_sot_token_id": 50361, "return_timestamps": false, "suppress_tokens": [1, 2, 7, 8, 9, 10, 14, 25, 26, 27, 28, 29, 31, 58, 59, 60, 61, 62, 63, 90, 91, 92, 93, 359, 503, 522, 542, 873, 893, 902, 918, 922, 931, 1350, 1853, 1982, 2460, 2627, 3246, 3253, 3268, 3536, 3846, 3961, 4183, 4667, 6585, 6647, 7273, 9061, 9383, 10428, 10929, 11938, 12033, 12331, 12562, 13793, 14157, 14635, 15265, 15618, 16553, 16604, 18362, 18956, 20075, 21675, 22520, 26130, 26161, 26435, 28279, 29464, 31650, 32302, 32470, 36865, 42863, 47425, 49870, 50254, 50258, 50358, 50359, 50360, 50361, 50362], "task_to_id": {"transcribe": 50359, "translate": 50358}, "transformers_version": "4.31.0.dev0"}

openai_whisper-small.en_217MB/AudioEncoder.mlmodelc/analytics/coremldata.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:45601c764f7bc51711b42670d55580fa949cf76bbeebc328c60882c048499bf2
-size 243

openai_whisper-small.en_217MB/AudioEncoder.mlmodelc/coremldata.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:a078e65c9369ce8a4a687a2bbb0a8befbd4ed459250c0442176824906fa95ee1
-size 433

openai_whisper-small.en_217MB/AudioEncoder.mlmodelc/metadata.json

@@ -1,92 +0,0 @@
-[
-  {
-    "metadataOutputVersion" : "3.0",
-    "storagePrecision" : "Mixed (Float16, Palettized (4 bits), Sparse)",
-    "outputSchema" : [
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 1 × 768 × 1 × 1500)",
-        "shortDescription" : "",
-        "shape" : "[1, 768, 1, 1500]",
-        "name" : "encoder_output_embeds",
-        "type" : "MultiArray"
-      },
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 12 × 768 × 1 × 1536)",
-        "shortDescription" : "",
-        "shape" : "[12, 768, 1, 1536]",
-        "name" : "encoder_attn_key_cache",
-        "type" : "MultiArray"
-      },
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 12 × 768 × 1 × 1536)",
-        "shortDescription" : "",
-        "shape" : "[12, 768, 1, 1536]",
-        "name" : "encoder_attn_value_cache",
-        "type" : "MultiArray"
-      }
-    ],
-    "modelParameters" : [
-
-    ],
-    "specificationVersion" : 9,
-    "mlProgramOperationTypeHistogram" : {
-      "Ios18.constexprLutToDense" : 98,
-      "Ios18.constexprSparseToDense" : 96,
-      "Ios18.conv" : 196,
-      "Ios18.matmul" : 24,
-      "Ios18.batchNorm" : 25,
-      "Pad" : 2,
-      "Ios18.gelu" : 14,
-      "Ios18.concat" : 2,
-      "Ios18.add" : 123,
-      "Ios18.softmax" : 12,
-      "Ios18.layerNorm" : 25,
-      "Ios18.reshape" : 48,
-      "Ios18.mul" : 12
-    },
-    "computePrecision" : "Mixed (Float16, Int32)",
-    "isUpdatable" : "0",
-    "stateSchema" : [
-
-    ],
-    "availability" : {
-      "macOS" : "15.0",
-      "tvOS" : "18.0",
-      "visionOS" : "2.0",
-      "watchOS" : "11.0",
-      "iOS" : "18.0",
-      "macCatalyst" : "18.0"
-    },
-    "modelType" : {
-      "name" : "MLModelType_mlProgram"
-    },
-    "userDefinedMetadata" : {
-      "com.github.apple.coremltools.source_dialect" : "TorchScript",
-      "com.github.apple.coremltools.source" : "torch==2.5.1",
-      "com.github.apple.coremltools.version" : "8.0"
-    },
-    "inputSchema" : [
-      {
-        "hasShapeFlexibility" : "0",
-        "isOptional" : "0",
-        "dataType" : "Float16",
-        "formattedType" : "MultiArray (Float16 1 × 80 × 1 × 3000)",
-        "shortDescription" : "",
-        "shape" : "[1, 80, 1, 3000]",
-        "name" : "melspectrogram_features",
-        "type" : "MultiArray"
-      }
-    ],
-    "generatedClassName" : "AudioEncoderStateful",
-    "method" : "predict"
-  }
-]

openai_whisper-small.en_217MB/AudioEncoder.mlmodelc/model.mlmodel

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:d08f8c4c63c48ed098e6536f5620273f3e661c17523e0861780a99d01a1a3749
-size 370371

openai_whisper-small.en_217MB/AudioEncoder.mlmodelc/weights/weight.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:265b53be8d21fd319531a81bf38db77112bfe068b7b681cbf47b26faccfbee55
-size 71137600

openai_whisper-small.en_217MB/LICENSE_NOTICE.txt

@@ -1,7 +0,0 @@
-Argmax proprietary and confidential. Under NDA.
-
-Copyright 2024 Argmax, Inc. All rights reserved.
-
-Unauthorized access, copying, use, distribution, and or commercialization of this file, via any medium or means is strictly prohibited.
-
-Please contact Argmax for licensing information at [email protected].

openai_whisper-small.en_217MB/MelSpectrogram.mlmodelc/analytics/coremldata.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:efc05e563ee0c556e3f578e04be5fb67b4e7520124403f2561f39102f0f2b33d
-size 243

openai_whisper-small.en_217MB/MelSpectrogram.mlmodelc/coremldata.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:e4ef11ea703011eab03287ec661f999e19c2c78cf67d531b5e6afa02e18f913d
-size 328