Medix-AI/coqui-tts
Medix-AI
/
coqui-tts
mirror of https://github.com/coqui-ai/TTS.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Fix glow tts default parameters

This commit is contained in:
Eren Gölge 2021-07-02 10:44:23 +02:00
parent a4c658f5ef
commit 0fa6a8c9b8
1 changed files with 7 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
TTS/tts/configs/glow_tts_config.py
View File

@ -41,7 +41,7 @@ class GlowTTSConfig(BaseTTSConfig):
kernel_size_dec (int): kernel_size_dec (int):
Decoder kernel size. Defaults to 5 Decoder kernel size. Defaults to 5
dilation_rate (int): dilation_rate (int):
Rate to increase dilation by each layer in a decoder block. Defaults to 5. Rate to increase dilation by each layer in a decoder block. Defaults to 1.
num_block_layers (int): num_block_layers (int):
Number of decoder layers in each decoder block. Defaults to 4. Number of decoder layers in each decoder block. Defaults to 4.
dropout_p_dec (float): dropout_p_dec (float):
@ -54,7 +54,7 @@ class GlowTTSConfig(BaseTTSConfig):
Number of split levels in inversible conv1x1 operation. Defaults to 4. Number of split levels in inversible conv1x1 operation. Defaults to 4.
num_squeeze (int): num_squeeze (int):
Number of squeeze levels. When squeezing channels increases and time steps reduces by the factor Number of squeeze levels. When squeezing channels increases and time steps reduces by the factor
'num_squeeze'. Defaults to 1. 'num_squeeze'. Defaults to 2.
sigmoid_scale (bool): sigmoid_scale (bool):
enable/disable sigmoid scaling in decoder. Defaults to False. enable/disable sigmoid scaling in decoder. Defaults to False.
mean_only (bool): mean_only (bool):
@ -74,6 +74,8 @@ class GlowTTSConfig(BaseTTSConfig):
Path to the wav file used for changing the style of the speech. Defaults to None. Path to the wav file used for changing the style of the speech. Defaults to None.
inference_noise_scale (float): inference_noise_scale (float):
Variance used for sampling the random noise added to the decoder's input at inference. Defaults to 0.0. Variance used for sampling the random noise added to the decoder's input at inference. Defaults to 0.0.
length_scale (float):
Multiply the predicted durations with this value to change the speech speed. Defaults to 1.
use_speaker_embedding (bool): use_speaker_embedding (bool):
enable / disable using speaker embeddings for multi-speaker models. If set True, the model is enable / disable using speaker embeddings for multi-speaker models. If set True, the model is
in the multi-speaker mode. Defaults to False. in the multi-speaker mode. Defaults to False.
@ -120,14 +122,13 @@ class GlowTTSConfig(BaseTTSConfig):
num_flow_blocks_dec: int = 12 num_flow_blocks_dec: int = 12
inference_noise_scale: float = 0.33 inference_noise_scale: float = 0.33
kernel_size_dec: int = 5 kernel_size_dec: int = 5
dilation_rate: int = 5 dilation_rate: int = 1
num_block_layers: int = 4 num_block_layers: int = 4
num_speakers: int = 0 num_speakers: int = 0
c_in_channels: int = 0 c_in_channels: int = 0
num_splits: int = 4 num_splits: int = 4
num_squeeze: int = 1 num_squeeze: int = 2
sigmoid_scale: bool = False sigmoid_scale: bool = False
mean_only: bool = False
encoder_type: str = "rel_pos_transformer" encoder_type: str = "rel_pos_transformer"
encoder_params: dict = field( encoder_params: dict = field(
default_factory=lambda: { default_factory=lambda: {
@ -147,6 +148,7 @@ class GlowTTSConfig(BaseTTSConfig):
# inference params # inference params
style_wav_for_test: str = None style_wav_for_test: str = None
inference_noise_scale: float = 0.0 inference_noise_scale: float = 0.0
length_scale: float = 1.0
# multi-speaker settings # multi-speaker settings
use_speaker_embedding: bool = False use_speaker_embedding: bool = False