File size: 4,708 Bytes
45b200f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
{
 "cells": [
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": "Audio",
   "id": "8b8c1a352260e82a"
  },
  {
   "cell_type": "code",
   "id": "initial_id",
   "metadata": {
    "collapsed": true,
    "ExecuteTime": {
     "end_time": "2025-06-10T09:38:10.760409Z",
     "start_time": "2025-06-10T09:38:10.617508Z"
    }
   },
   "source": [
    "from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor\n",
    "import torch\n",
    "import torchaudio.transforms as T\n",
    "import pydub\n",
    "import numpy as np"
   ],
   "outputs": [],
   "execution_count": 5
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-06-10T09:43:53.684713Z",
     "start_time": "2025-06-10T09:43:53.681866Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# CONSTANTS\n",
    "audio_model_dir = './models_for_proj/wav2vec2-base-960h'\n",
    "\n",
    "# audio_dir = 'files/1f975693-876d-457b-a649-393859e79bf3.mp3'\n",
    "audio_dir = 'files/99c9cc74-fdc8-46c6-8f8d-3ce2d3bfeea3.mp3'"
   ],
   "id": "3ee50d096b2c9d44",
   "outputs": [],
   "execution_count": 19
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-06-10T09:43:54.053411Z",
     "start_time": "2025-06-10T09:43:54.006676Z"
    }
   },
   "cell_type": "code",
   "source": [
    "\n",
    "model = Wav2Vec2ForCTC.from_pretrained(audio_model_dir)\n",
    "processor = Wav2Vec2Processor.from_pretrained(audio_model_dir)"
   ],
   "id": "b51a485af7b9cf14",
   "outputs": [],
   "execution_count": 20
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-06-10T09:43:54.603559Z",
     "start_time": "2025-06-10T09:43:54.414677Z"
    }
   },
   "cell_type": "code",
   "source": [
    "def read_mp3(f, normalized=False):\n",
    "    \"\"\"Read MP3 file to numpy array.\"\"\"\n",
    "    a = pydub.AudioSegment.from_mp3(f)\n",
    "    y = np.array(a.get_array_of_samples())\n",
    "    if a.channels == 2:\n",
    "        y = y.reshape((-1, 2))\n",
    "    if normalized:\n",
    "        return a.frame_rate, np.float32(y) / 2**15\n",
    "    else:\n",
    "        return a.frame_rate, y\n",
    "\n",
    "# Usage\n",
    "audio_input_sr, audio_input_np = read_mp3(audio_dir)"
   ],
   "id": "ac7e2b43ace4d232",
   "outputs": [],
   "execution_count": 21
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-06-10T09:43:56.920665Z",
     "start_time": "2025-06-10T09:43:56.244101Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# --------------------------------------------------------------------------- #\n",
    "# audio_input_sr, audio_input_np = audio_input\n",
    "audio_input_t = torch.tensor(audio_input_np, dtype=torch.float32)\n",
    "target_sr = 16000\n",
    "resampler = T.Resample(audio_input_sr, target_sr, dtype=audio_input_t.dtype)\n",
    "resampled_audio_input_t: torch.Tensor = resampler(audio_input_t)\n",
    "resampled_audio_input_np = resampled_audio_input_t.numpy()\n",
    "# --------------------------------------------------------------------------- #\n",
    "# result = asr_pipe_default(resampled_audio_input_np)\n",
    "inputs = processor(resampled_audio_input_np, sampling_rate=16000, return_tensors=\"pt\", padding=True)\n",
    "# Inference\n",
    "with torch.no_grad():\n",
    "    logits = model(**inputs).logits\n",
    "# Decode\n",
    "predicted_ids = torch.argmax(logits, dim=-1)\n",
    "transcription = processor.decode(predicted_ids[0])\n",
    "# print(\"Transcription:\", transcription)\n",
    "transcription"
   ],
   "id": "2a4738e9d038985",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'IN A SAUCEPAN COMBINE RIPE STRAWBERRIES GRANULATED SUGAR FRESHLY SQUEEZED LEMON JUICE AND CORNSTARCH COOK THE MIXTURE OF A MEDIUM HEAT STIRRING CONSTANTLY UNTIL IT THICKENS TO A SMOOTH CONSISTENCY REMOVE FROM HEAT AND STIR IN A DASH OF PURE VANILLA EXTRACT ALLOW THE STRAWBERRY PIE FEELING TO COOL BEFORE USING IT AS A DELICIOUS AND FRUITY FILLING FOR YOUR PIE CRUST'"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 22
  },
  {
   "metadata": {},
   "cell_type": "code",
   "outputs": [],
   "execution_count": null,
   "source": "",
   "id": "f159c2955f140600"
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}