class SpeakerEncoderConfig(BaseTrainingConfig):
"""Defines parameters for Speaker Encoder model."""
model: str = "speaker_encoder"
audio: BaseAudioConfig = field(default_factory=BaseAudioConfig)
datasets: List[BaseDatasetConfig] = field(
default_factory=lambda: [BaseDatasetConfig()])
# model params
model_params: Dict = field(
default_factory=lambda: {
"model_name": "lstm",
"input_dim": 80,
"proj_dim": 256,
"lstm_dim": 768,
"num_lstm_layers": 3,
"use_lstm_with_projection": True,
})
audio_augmentation: Dict = field(default_factory=lambda: {})
storage: Dict = field(
default_factory=lambda: {
"sample_from_storage_p":
0.66, # the probability with which we'll sample from the DataSet in-memory storage
"storage_size":
15, # the size of the in-memory storage with respect to a single batch
})
# training params
max_train_step: int = 1000000 # end training when number of training steps reaches this value.
loss: str = "angleproto"
grad_clip: float = 3.0
lr: float = 0.0001
lr_decay: bool = False
warmup_steps: int = 4000
wd: float = 1e-6
# logging params
tb_model_param_stats: bool = False
steps_plot_stats: int = 10
checkpoint: bool = True
save_step: int = 1000
print_step: int = 20
# data loader
num_speakers_in_batch: int = MISSING
num_utters_per_speaker: int = MISSING
num_loader_workers: int = MISSING
skip_speakers: bool = False
voice_len: float = 1.6
def check_values(self):
super().check_values()
c = asdict(self)
assert (
c["model_params"]["input_dim"] == self.audio.num_mels
), " [!] model input dimendion must be equal to melspectrogram dimension."
class BaseEncoderConfig(BaseTrainingConfig):
"""Defines parameters for a Generic Encoder model."""
model: str = None
audio: BaseAudioConfig = field(default_factory=BaseAudioConfig)
datasets: List[BaseDatasetConfig] = field(default_factory=lambda: [BaseDatasetConfig()])
# model params
model_params: Dict = field(
default_factory=lambda: {
"model_name": "lstm",
"input_dim": 80,
"proj_dim": 256,
"lstm_dim": 768,
"num_lstm_layers": 3,
"use_lstm_with_projection": True,
}
)
audio_augmentation: Dict = field(default_factory=lambda: {})
# training params
epochs: int = 10000
loss: str = "angleproto"
grad_clip: float = 3.0
lr: float = 0.0001
optimizer: str = "radam"
optimizer_params: Dict = field(default_factory=lambda: {"betas": [0.9, 0.999], "weight_decay": 0})
lr_decay: bool = False
warmup_steps: int = 4000
# logging params
tb_model_param_stats: bool = False
steps_plot_stats: int = 10
save_step: int = 1000
print_step: int = 20
run_eval: bool = False
# data loader
num_classes_in_batch: int = MISSING
num_utter_per_class: int = MISSING
eval_num_classes_in_batch: int = None
eval_num_utter_per_class: int = None
num_loader_workers: int = MISSING
voice_len: float = 1.6
def check_values(self):
super().check_values()
c = asdict(self)
assert (
c["model_params"]["input_dim"] == self.audio.num_mels
), " [!] model input dimendion must be equal to melspectrogram dimension."
import unittest
import torch
from tests import get_tests_output_path, run_cli
from TTS.config.shared_configs import BaseDatasetConfig
from TTS.tts.configs.vits_config import VitsConfig
torch.manual_seed(1)
config_path = os.path.join(get_tests_output_path(), "test_model_config.json")
dataset_config_en = BaseDatasetConfig(
name="ljspeech",
meta_file_train="metadata.csv",
meta_file_val="metadata.csv",
path="tests/data/ljspeech",
language="en",
)
dataset_config_pt = BaseDatasetConfig(
name="ljspeech",
meta_file_train="metadata.csv",
meta_file_val="metadata.csv",
path="tests/data/ljspeech",
language="pt-br",
)
# pylint: disable=protected-access
class TestFindUniquePhonemes(unittest.TestCase):
import os
from TTS.config.shared_configs import BaseAudioConfig, BaseDatasetConfig
from TTS.trainer import Trainer, TrainingArgs
from TTS.tts.configs.fast_pitch_config import FastPitchConfig
from TTS.tts.datasets import load_tts_samples
from TTS.tts.models.forward_tts import ForwardTTS
from TTS.utils.audio import AudioProcessor
from TTS.utils.manage import ModelManager
output_path = os.path.dirname(os.path.abspath(__file__))
# init configs
dataset_config = BaseDatasetConfig(
name="ljspeech",
meta_file_train="metadata.csv",
# meta_file_attn_mask=os.path.join(output_path, "../LJSpeech-1.1/metadata_attn_mask.txt"),
path=os.path.join(output_path, "../LJSpeech-1.1/"),
)
audio_config = BaseAudioConfig(
sample_rate=22050,
do_trim_silence=True,
trim_db=60.0,
signal_norm=False,
mel_fmin=0.0,
mel_fmax=8000,
spec_gain=1.0,
log_func="np.log",
ref_level_db=20,
preemphasis=0.0,
)