{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"id": "b8dcaaed",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Populating the interactive namespace from numpy and matplotlib\n"
]
}
],
"source": [
"%pylab inline"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "f60ee727",
"metadata": {},
"outputs": [],
"source": [
"import os\n",
"os.environ['CUDA_VISIBLE_DEVICES'] = ''"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "c50414bf",
"metadata": {},
"outputs": [],
"source": [
"from datetime import datetime, timedelta\n",
"import json\n",
"\n",
"import portion\n",
"\n",
"from suno_utils.utils.notebook import play_audio\n",
"\n",
"\n",
"def _convert_time_to_s(ts):\n",
" t = datetime.strptime(ts, \"%H:%M:%S.%f\")\n",
" td = timedelta(hours=t.hour, minutes=t.minute, seconds=t.second, microseconds=t.microsecond)\n",
" return td.total_seconds()"
]
},
{
"cell_type": "code",
"execution_count": 50,
"id": "ba566dff",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"7509\n"
]
}
],
"source": [
"DATA_DIR = \"/mnt/data-ssd-1/data/chime/CHiME5/\"\n",
"\n",
"transcripts = []\n",
"with open(DATA_DIR + \"transcriptions/dev/S02.json\") as f:\n",
" transcripts.extend(json.load(f))\n",
"with open(DATA_DIR + \"transcriptions/dev/S09.json\") as f:\n",
" transcripts.extend(json.load(f))\n",
"print(len(transcripts))"
]
},
{
"cell_type": "code",
"execution_count": 46,
"id": "74e4770b",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"298 [noise] [laughs] I was like, I was like, \"More country.\" [laughs]\n",
"300 If you can even get more country than College Station. It's like, \"You know, I like that station, but it's really not country enough for my taste.\" [laughs]\n",
"301 [laughs]\n",
"302 Do you want lemon in your water?\n",
"304 [laughs] That's like, uh\n"
]
}
],
"source": [
"# look for concurrent speeches\n",
"n = 300\n",
"base_meta = transcripts[n]\n",
"base_interval = portion.closedopen(\n",
" _convert_time_to_s(base_meta[\"start_time\"][\"original\"]), \n",
" _convert_time_to_s(base_meta[\"end_time\"][\"original\"]),\n",
")\n",
"for n, meta in enumerate(transcripts):\n",
" if meta[\"session_id\"] != \"S02\":\n",
" continue\n",
" check_interval = portion.closedopen(\n",
" _convert_time_to_s(meta[\"start_time\"][\"original\"]), \n",
" _convert_time_to_s(meta[\"end_time\"][\"original\"]),\n",
" )\n",
" overlap = (check_interval & base_interval)\n",
" if not overlap.empty and overlap.upper - overlap.lower >= 1:\n",
" print(n, meta[\"words\"])"
]
},
{
"cell_type": "code",
"execution_count": 47,
"id": "8bfb9086",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"If you can even get more country than College Station. It's like, \"You know, I like that station, but it's really not country enough for my taste.\" [laughs]\n"
]
},
{
"data": {
"text/html": [
"\n",
" \n",
" "
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"\n",
" \n",
" "
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"meta = transcripts[300]\n",
"print(meta[\"words\"])\n",
"# play lapel mic for speaker\n",
"play_audio(\n",
" DATA_DIR + \"audio/dev/\" + \"{}_{}.wav\".format(meta[\"session_id\"], meta[\"speaker\"]), \n",
" from_s=_convert_time_to_s(meta[\"start_time\"][meta[\"speaker\"]]), \n",
" to_s=_convert_time_to_s(meta[\"end_time\"][meta[\"speaker\"]]),\n",
")\n",
"# play one channel of the kinect that is in the correct room\n",
"play_audio(\n",
" DATA_DIR + \"audio/dev/\" + \"{}_{}.CH1.wav\".format(meta[\"session_id\"], meta[\"ref\"]), \n",
" from_s=_convert_time_to_s(meta[\"start_time\"][meta[\"ref\"]]), \n",
" to_s=_convert_time_to_s(meta[\"end_time\"][meta[\"ref\"]]),\n",
")"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "4cf5c1a9",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "c1b615f4",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "3ab851e5",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "ba2c1b75",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "c8a0d7fc",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "2084bbbb",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "22e91a07",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "a5981d79",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "b5307c63",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "fa4c68e0",
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.8.10"
}
},
"nbformat": 4,
"nbformat_minor": 5
}