deepgeek/CosyVoice
2
0
Fork 0
mirror of https://github.com/FunAudioLLM/CosyVoice.git synced 2026-02-05 09:59:23 +08:00
Code Issues Packages Projects Releases Wiki Activity

update 25hz yaml

Browse Source
This commit is contained in:
lyuxiang.lx
2024-09-29 10:42:57 +08:00
parent ffa28e3bbd
commit abc6f70ace
4 changed files with 22 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
examples/libritts/cosyvoice/conf/cosyvoice.fromscratch.yaml
View File

@@ -18,7 +18,7 @@ llm: !new:cosyvoice.llm.llm.TransformerLM
text_encoder_input_size: !ref <text_encoder_input_size>
llm_input_size: !ref <llm_input_size>
llm_output_size: !ref <llm_output_size>
text_token_size: 51866
text_token_size: 51866 # change to 60515 if you want to train with CosyVoice-300M-25Hz recipe
speech_token_size: 4096
length_normalized_loss: True
lsm_weight: 0
@@ -66,7 +66,7 @@ flow: !new:cosyvoice.flow.flow.MaskedDiffWithXvec
spk_embed_dim: !ref <spk_embed_dim>
output_type: 'mel'
vocab_size: 4096
input_frame_rate: 50
input_frame_rate: 50 # change to 25 if you want to train with CosyVoice-300M-25Hz recipe
only_mask_loss: True
encoder: !new:cosyvoice.transformer.encoder.ConformerEncoder
output_size: 512
@@ -135,7 +135,7 @@ hift: !new:cosyvoice.hifigan.generator.HiFTGenerator
# processor functions
parquet_opener: !name:cosyvoice.dataset.processor.parquet_opener
get_tokenizer: !name:whisper.tokenizer.get_tokenizer
get_tokenizer: !name:whisper.tokenizer.get_tokenizer # change to !name:cosyvoice.tokenizer.tokenizer.get_tokenizer if you want to train with CosyVoice-300M-25Hz recipe
multilingual: True
num_languages: 100
language: 'en'

6
examples/libritts/cosyvoice/conf/cosyvoice.yaml
View File

@@ -18,7 +18,7 @@ llm: !new:cosyvoice.llm.llm.TransformerLM
text_encoder_input_size: !ref <text_encoder_input_size>
llm_input_size: !ref <llm_input_size>
llm_output_size: !ref <llm_output_size>
text_token_size: 51866
text_token_size: 51866 # change to 60515 if you want to train with CosyVoice-300M-25Hz recipe
speech_token_size: 4096
length_normalized_loss: True
lsm_weight: 0
@@ -66,7 +66,7 @@ flow: !new:cosyvoice.flow.flow.MaskedDiffWithXvec
spk_embed_dim: !ref <spk_embed_dim>
output_type: 'mel'
vocab_size: 4096
input_frame_rate: 50
input_frame_rate: 50 # change to 25 if you want to train with CosyVoice-300M-25Hz recipe
only_mask_loss: True
encoder: !new:cosyvoice.transformer.encoder.ConformerEncoder
output_size: 512
@@ -135,7 +135,7 @@ hift: !new:cosyvoice.hifigan.generator.HiFTGenerator
# processor functions
parquet_opener: !name:cosyvoice.dataset.processor.parquet_opener
get_tokenizer: !name:whisper.tokenizer.get_tokenizer
get_tokenizer: !name:whisper.tokenizer.get_tokenizer # change to !name:cosyvoice.tokenizer.tokenizer.get_tokenizer if you want to train with CosyVoice-300M-25Hz recipe
multilingual: True
num_languages: 100
language: 'en'