{
"cells": [
{
"cell_type": "code",
"execution_count": 2,
"id": "3abd2681",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-25T14:31:14.201078Z",
"start_time": "2023-09-25T14:31:14.198314Z"
}
},
"outputs": [],
"source": [
"import os\n",
"os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"1\"\n",
"\n",
"from suno_utils.audio import Audio\n",
"from suno_utils.tasks.mert_v2 import (\n",
" encode as mert_encode,\n",
" preload_models as preload_mert_models\n",
")\n",
"from suno_utils.tasks.dac import (\n",
" encode as codec_encode,\n",
" decode as codec_decode,\n",
" preload_models as preload_codec_models,\n",
")\n",
"\n",
"from suno_utils.tasks.gpt_v2.chirp_v1 import preload_models as preload_chirp_models\n",
"from suno_utils.tasks.gpt_v2.chirp_v1 import generate_audio\n",
"from suno_utils.tasks.gpt_v2.generation import generate, load_model"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "08c0ad94",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-25T14:31:15.714611Z",
"start_time": "2023-09-25T14:31:14.500898Z"
}
},
"outputs": [],
"source": [
"%load_ext autoreload\n",
"%autoreload 2"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "f97f7e06",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-25T14:32:30.400466Z",
"start_time": "2023-09-25T14:31:15.715899Z"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"model loaded: 1720.9M params, 1.411 loss\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Some weights of the model checkpoint at m-a-p/MERT-v1-95M were not used when initializing MERTModel: ['encoder.pos_conv_embed.conv.weight_v', 'encoder.pos_conv_embed.conv.weight_g']\n",
"- This IS expected if you are initializing MERTModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
"- This IS NOT expected if you are initializing MERTModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
"Some weights of MERTModel were not initialized from the model checkpoint at m-a-p/MERT-v1-95M and are newly initialized: ['encoder.pos_conv_embed.conv.parametrizations.weight.original1', 'encoder.pos_conv_embed.conv.parametrizations.weight.original0']\n",
"You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
"/home/tony/anaconda3/envs/suno_env/lib/python3.10/site-packages/torch/nn/utils/weight_norm.py:30: UserWarning: torch.nn.utils.weight_norm is deprecated in favor of torch.nn.utils.parametrizations.weight_norm.\n",
" warnings.warn(\"torch.nn.utils.weight_norm is deprecated in favor of torch.nn.utils.parametrizations.weight_norm.\")\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Downloading s3://suno-data/georg/trained_models/chirp_v1/lid.176.bin to /tmp/tmpohwtj1vb\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Warning : `load_model` does not return WordVectorModel or SupervisedModel any more, but a `FastText` object which is very similar.\n"
]
}
],
"source": [
"preload_chirp_models(\n",
" centroids_filepath=\"/home/mikeys/bundle/2x1k_centroids_mert.npy\",\n",
" gpt_ckpt_path=\"/home/tony/Data/Chirp/xl_2.pt\",\n",
" codec_ckpt_path=\"/home/mikeys/bundle/d_codec_25x8.pt\",\n",
" common_genre_tags_path=\"/home/mikeys/bundle/common_tags.json\",\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "bf0de82d",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-25T14:33:33.898649Z",
"start_time": "2023-09-25T14:32:30.402010Z"
}
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2435/2435 [01:01<00:00, 39.37it/s]\n"
]
}
],
"source": [
"text = \"\"\"\n",
"[Verse]\n",
"In my kitchen, there's a problem I can't tame\n",
"The sink is stuck, it's driving me insane\n",
"Every dish piled up, I can't wash them away\n",
"This sticky situation ruins my day\n",
"\"\"\"\n",
"history_arr, raw_arrays, audios, text_tags = generate_audio(\n",
" text,\n",
" history_audio=None,\n",
" text_tags=\"[chinese, pop]\",\n",
" n_batch=4,\n",
" temp_semantic=0.95,\n",
" cfg_coef=1.25,\n",
" cfg_coef_tags=1.75,\n",
" max_history_duration_s=10.01,\n",
" max_gen_duration_s=30,\n",
" min_eos_p=0.25,\n",
" return_raw_arrays=True,\n",
" allow_genre_randomization=False,\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "cdcfc0dd",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-25T14:33:36.106277Z",
"start_time": "2023-09-25T14:33:33.900405Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"\n",
"\n"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"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": [
"for o in audios:\n",
" o.play()"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "15b0b2e5",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:07:23.935610Z",
"start_time": "2023-09-13T22:07:23.911467Z"
}
},
"outputs": [],
"source": [
"my_history=raw_arrays[0]"
]
},
{
"cell_type": "code",
"execution_count": 7,
"id": "09b6dd7a",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:07:23.989186Z",
"start_time": "2023-09-13T22:07:23.937333Z"
}
},
"outputs": [
{
"data": {
"text/plain": [
"array([[ 791, 1186, 530, ..., 2952, 581, 473],\n",
" [ 168, 4098, 4098, ..., 4098, 4098, 4098],\n",
" [ 979, 4098, 4098, ..., 4098, 4098, 4098],\n",
" ...,\n",
" [1000, 3522, 71, ..., 3744, 799, 2468],\n",
" [1000, 4098, 4098, ..., 4098, 4098, 4098],\n",
" [1000, 4098, 4098, ..., 4098, 4098, 4098]])"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"my_history"
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "43a4677e",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:07:24.044809Z",
"start_time": "2023-09-13T22:07:23.990526Z"
}
},
"outputs": [],
"source": [
"text2 = \"\"\"\n",
"[Chorus]\n",
"Oh, kitchen sink, won't you set me free\n",
"I despise this clog, it's my enemy\n",
"Won't someone save me from this misery\n",
"Oh, kitchen sink, release my agony\n",
"\"\"\""
]
},
{
"cell_type": "code",
"execution_count": 9,
"id": "8f4c620f",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:08:49.476338Z",
"start_time": "2023-09-13T22:07:24.046643Z"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"I can't wash them away The sticky situation who wins my day True\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████| 2435/2435 [01:18<00:00, 31.12it/s]\n"
]
}
],
"source": [
"history_arr_2, raw_arrays_2, audios_2, text_tags_2 = generate_audio(\n",
" text2,\n",
" history_audio=my_history,\n",
" text_tags=None,\n",
" n_batch=4,\n",
" temp_semantic=0.95,\n",
" cfg_coef=1.25,\n",
" cfg_coef_tags=1.75,\n",
" max_history_duration_s=10.01,\n",
" max_gen_duration_s=30,\n",
" min_eos_p=0.25,\n",
" return_raw_arrays=True,\n",
" allow_genre_randomization=False,\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 10,
"id": "cfd752f7",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:08:51.719002Z",
"start_time": "2023-09-13T22:08:49.477998Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"\n",
"\n"
],
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
},
{
"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": [
"for o in audios_2:\n",
" o.play()"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "91610776",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": 11,
"id": "3e785c62",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:08:51.985129Z",
"start_time": "2023-09-13T22:08:51.720217Z"
}
},
"outputs": [
{
"ename": "NameError",
"evalue": "name 'BREAK' is not defined",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)",
"Cell \u001b[0;32mIn[11], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m \u001b[43mBREAK\u001b[49m\n",
"\u001b[0;31mNameError\u001b[0m: name 'BREAK' is not defined"
]
}
],
"source": [
"BREAK"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "edc90576",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:08:51.986670Z",
"start_time": "2023-09-13T22:08:51.986661Z"
}
},
"outputs": [],
"source": [
"text = \"\"\"\n",
"[Verse]\n",
"一壶漂泊浪迹天涯难入喉\n",
"你走之后酒暖回忆思念瘦\n",
"水向东流时间怎么偷\n",
"花开就一次成熟我却错过\n",
"[Chorus]\n",
"谁在用琵琶弹奏一曲东风破\n",
"枫叶将故事染色结局我看透\n",
"篱笆外的古道我牵着你走过\n",
"荒烟蔓草的年头就连分手都很沉默\n",
"\"\"\"\n",
"out = generate_audio(\n",
" text,\n",
" text_tags=\"[chinese, pop]\",\n",
" n_batch=4,\n",
" # cfg_coef=4.0,\n",
" # cfg_coef_tags_max_steps=100,\n",
" max_gen_duration_s=30,\n",
" allow_eos=True,\n",
")"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "0848e99e",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:08:51.987532Z",
"start_time": "2023-09-13T22:08:51.987523Z"
}
},
"outputs": [],
"source": [
"for o in out:\n",
" o.play()"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "34ae5b1d",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:08:51.988273Z",
"start_time": "2023-09-13T22:08:51.988263Z"
}
},
"outputs": [],
"source": [
"out_2 = generate_audio(\n",
" text,\n",
" text_tags=\"[chinese, pop]\",\n",
" n_batch=4,\n",
" # cfg_coef=4.0,\n",
" # cfg_coef_tags_max_steps=100,\n",
" max_gen_duration_s=40,\n",
" allow_eos=True,\n",
")"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "0f700002",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:08:51.988989Z",
"start_time": "2023-09-13T22:08:51.988980Z"
}
},
"outputs": [],
"source": [
"for o in out_2:\n",
" o.play()"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "38b81107",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:08:51.989648Z",
"start_time": "2023-09-13T22:08:51.989640Z"
}
},
"outputs": [],
"source": [
"STOP"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "456f215b",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:08:51.990387Z",
"start_time": "2023-09-13T22:08:51.990378Z"
}
},
"outputs": [],
"source": [
"# _ = generate_audio(\n",
"# text=\"blabla\",\n",
"# text_tags=\"blabla\",\n",
"# n_batch=2,\n",
"# max_gen_duration_s=30,\n",
"# allow_eos=False,\n",
"# )"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "cceb2397",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:08:51.991085Z",
"start_time": "2023-09-13T22:08:51.991076Z"
}
},
"outputs": [],
"source": [
"text = \"\"\"\n",
"\\n\n",
"\\n\n",
"\\n\n",
"\\n\n",
"[Verse]\n",
"In my kitchen, there's a problem I can't tame\n",
"The sink is stuck, it's driving me insane\n",
"Every dish piled up, I can't wash them away\n",
"This sticky situation ruins my day\n",
"\n",
"[Chorus]\n",
"Oh, kitchen sink, won't you set me free\n",
"I despise this clog, it's my enemy\n",
"Won't someone save me from this misery\n",
"Oh, kitchen sink, release my agony\n",
"\\n\n",
"\\n\n",
"\\n\n",
"\\n\n",
"\"\"\"\n",
"out = generate_audio(\n",
" text,\n",
" text_tags=\"[kpop]\", \n",
" n_batch=2,\n",
" history_audio=None,\n",
" history_text=None,\n",
" max_gen_duration_s=40,\n",
" max_history_duration_s=10.01,\n",
" temp_semantic=0.85,\n",
" temp_coarse=0.8,\n",
" top_k_semantic=500,\n",
" top_k_coarse=500,\n",
" cfg_coef=1.25,\n",
" cfg_coef_tags=1.75,\n",
" min_eos_p=0.1,\n",
" eos_pad_duration_s=0.12,\n",
" rep_penality=0.2,\n",
" allow_genre_randomization=False,\n",
" silent=False,\n",
" return_raw_arrays = True\n",
")"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "f1cbb567",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:08:51.991834Z",
"start_time": "2023-09-13T22:08:51.991825Z"
}
},
"outputs": [],
"source": [
"#codec_decode(out[1][1][0]).play()\n",
"for o in out[2]:\n",
" o.play()\n",
"# or out[0][0].play()"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "8702302e",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "c754f2d7",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:08:51.992522Z",
"start_time": "2023-09-13T22:08:51.992513Z"
}
},
"outputs": [],
"source": [
"BREAK"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "e3c7d909",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:08:51.993138Z",
"start_time": "2023-09-13T22:08:51.993129Z"
}
},
"outputs": [],
"source": [
"my_audio = o"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "e9572f6c",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:08:51.993952Z",
"start_time": "2023-09-13T22:08:51.993943Z"
}
},
"outputs": [],
"source": [
"text_2 = \"\"\"\n",
"[Chorus]\n",
"Oh, kitchen sink, won't you set me free\n",
"I despise this clog, it's my enemy\n",
"Won't someone save me from this misery\n",
"Oh, kitchen sink, release my agony\n",
"\n",
"[Beat]\n",
"oh la\n",
"[Beat]\n",
"\"\"\"\n",
"out_2 = generate_audio(\n",
" text_2,\n",
" text_tags=\"[kpop]\", \n",
" n_batch=2,\n",
" history_audio=my_audio,\n",
" history_text=None,\n",
" max_gen_duration_s=30,\n",
" max_history_duration_s=10.01,\n",
" temp_semantic=0.85,\n",
" temp_coarse=0.8,\n",
" top_k_semantic=250,\n",
" top_k_coarse=500,\n",
" cfg_coef=1.3,\n",
" cfg_coef_tags=1.75,\n",
" min_eos_p=0.1,\n",
" rep_penality=0.2,\n",
" allow_genre_randomization=False,\n",
" silent=False,\n",
" return_raw_arrays = True\n",
")"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "53f52760",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:08:51.994634Z",
"start_time": "2023-09-13T22:08:51.994625Z"
}
},
"outputs": [],
"source": [
"for o2 in out_2[2]:\n",
" o2.play()"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "7bc0c1ca",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:08:51.995325Z",
"start_time": "2023-09-13T22:08:51.995316Z"
}
},
"outputs": [],
"source": [
"out_3 = generate_audio(\n",
" text_2,\n",
" text_tags=\"[kpop]\", \n",
" n_batch=2,\n",
" history_audio=None,\n",
" history_text=None,\n",
" max_gen_duration_s=30,\n",
" max_history_duration_s=10.01,\n",
" temp_semantic=0.85,\n",
" temp_coarse=0.8,\n",
" top_k_semantic=250,\n",
" top_k_coarse=500,\n",
" cfg_coef=1.3,\n",
" cfg_coef_tags=1.75,\n",
" min_eos_p=0.1,\n",
" rep_penality=0.2,\n",
" allow_genre_randomization=False,\n",
" silent=False,\n",
" return_raw_arrays = True,\n",
" oracle_arr=out[1][1][:, 0].reshape(-1, 1)\n",
")"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "d2e5ddd9",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:08:51.996023Z",
"start_time": "2023-09-13T22:08:51.996014Z"
}
},
"outputs": [],
"source": [
"out[1][1][:, 0].reshape(-1, 1).shape"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "0393f011",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:08:51.996711Z",
"start_time": "2023-09-13T22:08:51.996702Z"
}
},
"outputs": [],
"source": [
"for o3 in out_3[2]:\n",
" o3.play()"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "557c57b0",
"metadata": {
"ExecuteTime": {
"end_time": "2023-09-13T22:08:51.997367Z",
"start_time": "2023-09-13T22:08:51.997358Z"
}
},
"outputs": [],
"source": [
"# text_2 = \"\"\"\n",
"# [Verse]\n",
"# 我在写一些新的\n",
"# 动听的音乐\n",
"# 让我们都陶醉其中\n",
"# 跳动的旋律\n",
"# 沁入心底\n",
"# 每一拍都让我们迷醉\n",
"# [Chorus]\n",
"# 大家很喜欢，我们疯狂沉迷\n",
"# 奔放的节奏让我们欢喜\n",
"# 抛开所有烦恼 追随梦想\n",
"# 这首歌永远都不会停\n",
"\n",
"# [Verse]\n",
"# 继续创作新的音乐，令人陶醉\n",
"# 在这旅途上勇往直前，毫不犹豫\n",
"# 音符飞舞翩跹，心灵荡漾，自由的心情\n",
"# 用激情唤醒每个人的感动\n",
"\n",
"# [Chorus]\n",
"# 继续往前走，我们不停舞蹈\n",
"# 向前再出发，奏响缤纷的色彩\n",
"# 绽放我们的梦想，让世界为之沉醉\n",
"# 唤起奋斗的力量，永远不停歇\n",
"\n",
"# [Beat]\n",
"# oh la\n",
"# [Beat]\n",
"# \"\"\"\n",
"# out_2 = generate_audio(\n",
"# text_2,\n",
"# text_tags=\"[mandopop][kpop]\", \n",
"# n_batch=2,\n",
"# history_audio=my_audio,\n",
"# history_text=text,\n",
"# max_gen_duration_s=30,\n",
"# max_history_duration_s=10.01,\n",
"# temp_semantic=0.85,\n",
"# temp_coarse=0.8,\n",
"# top_k_semantic=250,\n",
"# top_k_coarse=500,\n",
"# cfg_coef=1.3,\n",
"# cfg_coef_tags=1.75,\n",
"# min_eos_p=0.1,\n",
"# rep_penality=0.2,\n",
"# allow_genre_randomization=False,\n",
"# silent=False,\n",
"# )"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "3e3daf18",
"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": "3.10.12"
},
"toc": {
"base_numbering": 1,
"nav_menu": {},
"number_sections": true,
"sideBar": true,
"skip_h1_title": false,
"title_cell": "Table of Contents",
"title_sidebar": "Contents",
"toc_cell": false,
"toc_position": {},
"toc_section_display": true,
"toc_window_display": false
}
},
"nbformat": 4,
"nbformat_minor": 5
}