Added new model trained on 60 epochs

.ipynb_checkpoints/ASR_Inference-checkpoint.ipynb

@@ -2,23 +2,14 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 15,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T09:33:41.892030Z",
-     "start_time": "2021-03-14T09:33:40.729163Z"
+     "end_time": "2021-03-14T18:07:15.328900Z",
+     "start_time": "2021-03-14T18:07:15.326838Z"
     }
    },
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/home/earendil/anaconda3/envs/cuda110/lib/python3.8/site-packages/torchaudio/backend/utils.py:53: UserWarning: \"sox\" backend is being deprecated. The default backend will be changed to \"sox_io\" backend in 0.8.0 and \"sox\" backend will be removed in 0.9.0. Please migrate to \"sox_io\" backend. Please refer to https://github.com/pytorch/audio/issues/903 for the detail.\n",
-      "  warnings.warn(\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "from transformers import Wav2Vec2ForCTC\n",
     "from transformers import Wav2Vec2Processor\n",
@@ -33,11 +24,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 16,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T09:33:41.909851Z",
-     "start_time": "2021-03-14T09:33:41.906327Z"
+     "end_time": "2021-03-14T18:07:15.933957Z",
+     "start_time": "2021-03-14T18:07:15.927789Z"
     }
    },
    "outputs": [],
@@ -75,11 +66,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 17,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T09:33:49.053762Z",
-     "start_time": "2021-03-14T09:33:41.922683Z"
+     "end_time": "2021-03-14T18:07:22.624226Z",
+     "start_time": "2021-03-14T18:07:16.402381Z"
     }
    },
    "outputs": [
@@ -92,17 +83,17 @@
     }
    ],
    "source": [
-    "model = Wav2Vec2ForCTC.from_pretrained(\"wav2vec2-large-xlsr-greek/checkpoint-9200/\").to(\"cuda\")\n",
+    "model = Wav2Vec2ForCTC.from_pretrained(\"wav2vec2-large-xlsr-greek/checkpoint-18400/\").to(\"cuda\")\n",
     "processor = Wav2Vec2Processor.from_pretrained(\"wav2vec2-large-xlsr-greek/\")"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 18,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T09:33:52.413558Z",
-     "start_time": "2021-03-14T09:33:49.078466Z"
+     "end_time": "2021-03-14T18:07:25.473609Z",
+     "start_time": "2021-03-14T18:07:22.644765Z"
     }
    },
    "outputs": [
@@ -121,11 +112,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 19,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T09:33:52.444418Z",
-     "start_time": "2021-03-14T09:33:52.441338Z"
+     "end_time": "2021-03-14T18:07:25.504511Z",
+     "start_time": "2021-03-14T18:07:25.500688Z"
     }
    },
    "outputs": [],
@@ -135,11 +126,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 20,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T09:33:52.473087Z",
-     "start_time": "2021-03-14T09:33:52.468014Z"
+     "end_time": "2021-03-14T18:07:25.540666Z",
+     "start_time": "2021-03-14T18:07:25.536214Z"
     }
    },
    "outputs": [
@@ -157,11 +148,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 21,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T09:33:52.510377Z",
-     "start_time": "2021-03-14T09:33:52.501677Z"
+     "end_time": "2021-03-14T18:07:25.578015Z",
+     "start_time": "2021-03-14T18:07:25.568808Z"
     }
    },
    "outputs": [
@@ -179,11 +170,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 22,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T09:33:53.321810Z",
-     "start_time": "2021-03-14T09:33:52.533233Z"
+     "end_time": "2021-03-14T18:07:26.404914Z",
+     "start_time": "2021-03-14T18:07:25.605177Z"
     }
    },
    "outputs": [
@@ -191,30 +182,28 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "  "
+      " "
      ]
     },
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-ba8c6dd59eb8ccf2.arrow\n",
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-2e240883a5f827fd.arrow\n"
+      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-ba8c6dd59eb8ccf2.arrow\n"
      ]
     },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "  "
+      " "
      ]
     },
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-485c00dc9048ed50.arrow\n",
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-44bf1791baae8e2e.arrow\n"
+      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-2e240883a5f827fd.arrow\n"
      ]
     },
     {
@@ -228,35 +217,37 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-ecc0dfac5615a58e.arrow\n"
+      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-485c00dc9048ed50.arrow\n"
      ]
     },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      " "
+      "  "
      ]
     },
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-923d905502a8661d.arrow\n"
+      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-44bf1791baae8e2e.arrow\n",
+      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-ecc0dfac5615a58e.arrow\n"
      ]
     },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      " "
+      "  "
      ]
     },
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-062aeafc3b8816c1.arrow\n"
+      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-bb54bb00dae79669.arrow\n",
+      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-923d905502a8661d.arrow\n"
      ]
     },
     {
@@ -270,7 +261,7 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-bb54bb00dae79669.arrow\n"
+      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-062aeafc3b8816c1.arrow\n"
      ]
     }
    ],
@@ -280,62 +271,27 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 23,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T09:33:53.611415Z",
-     "start_time": "2021-03-14T09:33:53.342487Z"
+     "end_time": "2021-03-14T18:07:27.032511Z",
+     "start_time": "2021-03-14T18:07:26.432613Z"
     }
    },
    "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/home/earendil/anaconda3/envs/cuda110/lib/python3.8/site-packages/numpy/core/_asarray.py:83: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray\n",
-      "  return array(a, dtype, copy=False, order=order)\n"
-     ]
-    },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      " "
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-6dfad29ca815f865.arrow\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      " "
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-61e9ae0296df46f8.arrow\n"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      " "
+      "  "
      ]
     },
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-7f5aae16804e0788.arrow\n"
+      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-82be72eab73488a6.arrow\n",
+      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-a30edec53656694c.arrow\n"
      ]
     },
     {
@@ -349,7 +305,7 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-b9636a5d30ffb973.arrow\n"
+      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-91aacc366ff3e776.arrow\n"
      ]
     },
     {
@@ -363,7 +319,7 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-7e60f2d73a65610a.arrow\n"
+      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-cce8223f5c38f863.arrow\n"
      ]
     },
     {
@@ -377,21 +333,22 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-3c99781789816a60.arrow\n"
+      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-4f0d5b132b7516de.arrow\n"
      ]
     },
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      " "
+      "  "
      ]
     },
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-bae077f32f9eb290.arrow\n"
+      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-55caed3924d51e22.arrow\n",
+      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-2cc086daed2595be.arrow\n"
      ]
     },
     {
@@ -405,7 +362,7 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-4fb6951626f7548e.arrow\n"
+      "Loading cached processed dataset at /home/earendil/.cache/huggingface/datasets/common_voice/el-afd0a157f05ee080/6.1.0/32954a9015faa0d840f6c6894938545c5d12bc5d8936a80079af74bf50d71564/cache-118401c99df7b83c.arrow\n"
      ]
     }
    ],
@@ -415,11 +372,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 24,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T09:33:56.243678Z",
-     "start_time": "2021-03-14T09:33:53.632436Z"
+     "end_time": "2021-03-14T18:07:29.428864Z",
+     "start_time": "2021-03-14T18:07:27.056686Z"
     }
    },
    "outputs": [
@@ -438,17 +395,17 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 25,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T09:36:50.076837Z",
-     "start_time": "2021-03-14T09:36:24.943947Z"
+     "end_time": "2021-03-14T18:07:54.722520Z",
+     "start_time": "2021-03-14T18:07:29.451275Z"
     }
    },
    "outputs": [],
    "source": [
     "# Change this value to try inference on different CommonVoice extracts\n",
-    "example = 123\n",
+    "example = 678\n",
     "\n",
     "input_dict = processor(common_voice_test[\"input_values\"][example], return_tensors=\"pt\", sampling_rate=16_000, padding=True)\n",
     "\n",
@@ -459,11 +416,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 26,
    "metadata": {
     "ExecuteTime": {
-     "end_time": "2021-03-14T09:36:50.137886Z",
-     "start_time": "2021-03-14T09:36:50.134218Z"
+     "end_time": "2021-03-14T18:07:54.742988Z",
+     "start_time": "2021-03-14T18:07:54.739626Z"
     }
    },
    "outputs": [
@@ -472,29 +429,22 @@
      "output_type": "stream",
      "text": [
       "Prediction:\n",
-      "καμιά φορά τα έπαιρνε και έπαιζε όταν η δουλειά ήταν πιο χαλαρί\n",
+      "πού θέλεις να πάμε ρώτησε φοβισμένα ο βασιλιάς\n",
       "\n",
       "Reference:\n",
-      "καμιά φορά τα έπαιρνε και έπαιζε όταν η δουλειά ήταν πιο χαλαρή\n"
+      "πού θέλεις να πάμε; ρώτησε φοβισμένα ο βασιλιάς.\n"
      ]
     }
    ],
    "source": [
     "print(\"Prediction:\")\n",
     "print(processor.decode(pred_ids[0]))\n",
-    "# καμιά φορά τα έπαιρνε και έπαιζε όταν η δουλειά ήταν πιο χαλαρί\n",
+    "# πού θέλεις να πάμε ρώτησε φοβισμένα ο βασιλιάς\n",
     "\n",
     "print(\"\\nReference:\")\n",
     "print(common_voice_test_transcription[\"sentence\"][example].lower())\n",
-    "# καμιά φορά τα έπαιρνε και έπαιζε όταν η δουλειά ήταν πιο χαλαρή"
+    "# πού θέλεις να πάμε; ρώτησε φοβισμένα ο βασιλιάς."
    ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": []
   }
  ],
  "metadata": {