mirrors/icefall
1
0
Fork 0
mirror of https://github.com/k2-fsa/icefall.git synced 2025-08-09 10:02:22 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'k2-fsa:master' into dev/speechllm

Browse Source
This commit is contained in:
Yifan Yang 2025-05-24 13:13:42 +08:00 committed by GitHub
commit d1a535dc76
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
3 changed files with 8 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
README.md
View File

@ -383,3 +383,7 @@ Please see: [![Open In Colab](https://colab.research.google.com/assets/colab-bad
[vctk]: egs/vctk/TTS [vctk]: egs/vctk/TTS
[ljspeech]: egs/ljspeech/TTS [ljspeech]: egs/ljspeech/TTS
[libritts_tts]: egs/libritts/TTS [libritts_tts]: egs/libritts/TTS
## Acknowledgements
Some contributors to this project were supported by Xiaomi Corporation. Others were supported by National Science Foundation CCRI award 2120435. This is not an exhaustive list of sources of support.

4
egs/librispeech/ASR/zipformer/finetune.py
View File

@ -140,8 +140,8 @@ def add_finetune_arguments(parser: argparse.ArgumentParser):
type=str2bool, type=str2bool,
default=False, default=False,
help=""" help="""
Whether to adapt. If true, we will mix 5% of the new data Whether to adapt. If true, we will mix 5%% of the new data
with 95% of the original data to fine-tune. This is useful with 95%% of the original data to fine-tune. This is useful
if you want to maintain the performance on the original domain if you want to maintain the performance on the original domain
""", """,
) )

2
icefall/utils.py
View File

@ -186,7 +186,7 @@ class AttributeDict(dict):
tmp = {} tmp = {}
for k, v in self.items(): for k, v in self.items():
# PosixPath is ont JSON serializable # PosixPath is ont JSON serializable
if isinstance(v, pathlib.Path) or isinstance(v, torch.device): if isinstance(v, (pathlib.Path, torch.device, torch.dtype)):
v = str(v) v = str(v)
tmp[k] = v tmp[k] = v
return json.dumps(tmp, indent=indent, sort_keys=True) return json.dumps(tmp, indent=indent, sort_keys=True)