Making training more generic

This commit is contained in:
Dave Niewinski 2024-02-14 11:02:28 -05:00
parent d67690c1eb
commit 855a69cfdd

View File

@ -13,6 +13,24 @@
"#!pip3 install nemo_toolkit[all]" "#!pip3 install nemo_toolkit[all]"
] ]
}, },
{
"cell_type": "code",
"execution_count": null,
"id": "ab5977d6-964e-4ebf-bbd8-e5418fdf67ea",
"metadata": {},
"outputs": [],
"source": [
"## GLaDOS ##\n",
"blocklist = [\"potato\", \"_ding_\", \"00_part1_entry-6\", \"_escape_\"]\n",
"sources = [\"https://theportalwiki.com/wiki/GLaDOS_voice_lines_(Portal)\", \"https://theportalwiki.com/wiki/GLaDOS_voice_lines_(Portal_2)\", \"https://theportalwiki.com/wiki/GLaDOS_voice_lines_(Other)\"]\n",
"base_dir=\"glados_out\"\n",
"\n",
"## Wheatley ##\n",
"#blocklist = [\"jailbreakdooropens08\", \"jailbreakdooropens05\", \"jailbreakdooropens07\", \"jailbreakdooropens01\", \"sp_a1_intro7_pickupnags07\", \"sp_a1_wakeup_panic01\", \"sp_a1_wakeup_hacking09\", \"sp_a1_wakeup_hacking12\", \"sp_a1_wakeup_hacking10\", \"sp_a1_wakeup_hacking12\", \"sp_trust_flingalt08\", \"nanobotow03\", \"sp_a2_wheatley_ows\", \"bw_a4_2nd_first_test_solve_nags\", \"bw_sp_a2_core_actually05\", \"bw_sp_a2_core_actually01\", \"bw_sp_a2_core_potato04\", \"bw_sp_a4_tb_trust_drop_solve05\", \"bw_sp_a4_tb_trust_drop_impatient02\", \"bw_sp_a4_tb_wall_button_solve09\", \"bw_sp_a4_tb_polarity_solve04\", \"bw_a4_finale04_whitegel_break01\", \"bw_finale04_portal_opens_short05\", \"bw_finale04_portal_opens_short06\", \"bw_finale04_portal_opens_short07\", \"bw_finale04_portal_opens_short08\", \"bw_a4_finale04_wakeupa09\", \"bw_finale04_portal_opens17\", \"bw_finale04_portal_opens15\", \"bw_finale04_portal_opens13\", \"bw_finale04_portal_opens03\", \"openingwallhitone01\", \"demospherepowerup02\"]\n",
"#sources = [\"https://theportalwiki.com/wiki/wheatley_voice_lines\"]\n",
"#base_dir=\"wheatley_out\"\n"
]
},
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null, "execution_count": null,
@ -75,8 +93,6 @@
" BOLD = '\\033[1m'\n", " BOLD = '\\033[1m'\n",
" UNDERLINE = '\\033[4m'\n", " UNDERLINE = '\\033[4m'\n",
"\n", "\n",
"blocklist = [\"jailbreakdooropens08\", \"jailbreakdooropens05\", \"jailbreakdooropens07\", \"jailbreakdooropens01\"]\n",
"\n",
"audio_dir = 'audio'\n", "audio_dir = 'audio'\n",
"download_threads = 64\n", "download_threads = 64\n",
"\n", "\n",
@ -127,9 +143,6 @@
" print(bcolors.FAIL + \"[\" + u'\\u2715' + \"] \" + bcolors.ENDC + result[0])\n", " print(bcolors.FAIL + \"[\" + u'\\u2715' + \"] \" + bcolors.ENDC + result[0])\n",
"\n", "\n",
"def main():\n", "def main():\n",
" sources = [\"https://theportalwiki.com/wiki/wheatley_voice_lines\"]\n",
" #sources = [\"https://theportalwiki.com/wiki/GLaDOS_voice_lines_(Portal)\", \"https://theportalwiki.com/wiki/GLaDOS_voice_lines_(Portal_2)\", \"https://theportalwiki.com/wiki/GLaDOS_voice_lines_(Other)\"]\n",
" \n",
" urls = []\n", " urls = []\n",
" filenames = []\n", " filenames = []\n",
" texts = []\n", " texts = []\n",
@ -280,11 +293,11 @@
"outputs": [], "outputs": [],
"source": [ "source": [
"# additional files\n", "# additional files\n",
"!mkdir -p tts_dataset_files && cd tts_dataset_files \\\n", "#!mkdir -p tts_dataset_files && cd tts_dataset_files \\\n",
"&& wget https://raw.githubusercontent.com/NVIDIA/NeMo/r1.23.0/scripts/tts_dataset_files/cmudict-0.7b_nv22.10 \\\n", "#&& wget https://raw.githubusercontent.com/NVIDIA/NeMo/r1.23.0/scripts/tts_dataset_files/cmudict-0.7b_nv22.10 \\\n",
"&& wget https://raw.githubusercontent.com/NVIDIA/NeMo/r1.23.0/scripts/tts_dataset_files/heteronyms-052722 \\\n", "#&& wget https://raw.githubusercontent.com/NVIDIA/NeMo/r1.23.0/scripts/tts_dataset_files/heteronyms-052722 \\\n",
"&& wget https://raw.githubusercontent.com/NVIDIA/NeMo/r1.23.0/scripts/dataset_processing/tts/ljspeech/lj_speech.tsv \\\n", "#&& wget https://raw.githubusercontent.com/NVIDIA/NeMo/r1.23.0/scripts/dataset_processing/tts/ljspeech/lj_speech.tsv \\\n",
"&& cd .." "#&& cd .."
] ]
}, },
{ {
@ -362,8 +375,8 @@
" sup_data_path=./fastpitch_sup_data \\\n", " sup_data_path=./fastpitch_sup_data \\\n",
" phoneme_dict_path=tts_dataset_files/cmudict-0.7b_nv22.10 \\\n", " phoneme_dict_path=tts_dataset_files/cmudict-0.7b_nv22.10 \\\n",
" heteronyms_path=tts_dataset_files/heteronyms-052722 \\\n", " heteronyms_path=tts_dataset_files/heteronyms-052722 \\\n",
" +whitelist_path=tts_dataset_files/lj_speech.tsv \\\n", "# +whitelist_path=tts_dataset_files/lj_speech.tsv \\\n",
" exp_manager.exp_dir=./wheatley_out \\\n", " exp_manager.exp_dir=./{base_dir} \\\n",
" +init_from_nemo_model=./tts_en_fastpitch_align.nemo \\\n", " +init_from_nemo_model=./tts_en_fastpitch_align.nemo \\\n",
" trainer.max_epochs=300 \\\n", " trainer.max_epochs=300 \\\n",
" trainer.check_val_every_n_epoch=25 \\\n", " trainer.check_val_every_n_epoch=25 \\\n",
@ -430,7 +443,7 @@
" return spectrogram, audio\n", " return spectrogram, audio\n",
"\n", "\n",
"def get_best_ckpt_from_last_run(\n", "def get_best_ckpt_from_last_run(\n",
" base_dir=\"./wheatley_out\", \n", " base_dir=\"./\" + base_dir, \n",
" model_name=\"FastPitch\"\n", " model_name=\"FastPitch\"\n",
" ): \n", " ): \n",
" \n", " \n",
@ -589,7 +602,7 @@
" ~model.optim.sched \\\n", " ~model.optim.sched \\\n",
" train_dataset=./hifigan_manifest_train.json \\\n", " train_dataset=./hifigan_manifest_train.json \\\n",
" validation_datasets=./hifigan_manifest_validation.json \\\n", " validation_datasets=./hifigan_manifest_validation.json \\\n",
" exp_manager.exp_dir=./wheatley_out \\\n", " exp_manager.exp_dir=./{base_dir} \\\n",
" +init_from_pretrained_model=tts_en_hifigan \\\n", " +init_from_pretrained_model=tts_en_hifigan \\\n",
" trainer.check_val_every_n_epoch=10 \\\n", " trainer.check_val_every_n_epoch=10 \\\n",
" model/train_ds=train_ds_finetune \\\n", " model/train_ds=train_ds_finetune \\\n",
@ -613,8 +626,6 @@
"# Only need to set speaker_id if there is more than one speaker\n", "# Only need to set speaker_id if there is more than one speaker\n",
"speaker_id = None\n", "speaker_id = None\n",
"\n", "\n",
"base_dir=\"wheatley_out\"\n",
"\n",
"last_fastpitch_ckpt = get_best_ckpt_from_last_run(base_dir=base_dir, model_name=\"FastPitch\")\n", "last_fastpitch_ckpt = get_best_ckpt_from_last_run(base_dir=base_dir, model_name=\"FastPitch\")\n",
"last_hifigan_ckpt = get_best_ckpt_from_last_run(base_dir=base_dir, model_name=\"HifiGan\")\n", "last_hifigan_ckpt = get_best_ckpt_from_last_run(base_dir=base_dir, model_name=\"HifiGan\")\n",
"\n", "\n",
@ -671,6 +682,10 @@
"id": "9706d079-fd76-49f6-bce9-1a902d930699", "id": "9706d079-fd76-49f6-bce9-1a902d930699",
"metadata": {}, "metadata": {},
"source": [ "source": [
"#### NAMING WILL CHANGE DEPENDING ON CHARACTER ####\n",
"#### NAMING WILL CHANGE DEPENDING ON CHARACTER ####\n",
"#### NAMING WILL CHANGE DEPENDING ON CHARACTER ####\n",
"\n",
"# pip3 install whl\n", "# pip3 install whl\n",
"# pip3 install nemo2riva\n", "# pip3 install nemo2riva\n",
"\n", "\n",