mirrors/icefall
1
0
Fork 0
mirror of https://github.com/k2-fsa/icefall.git synced 2025-08-26 18:24:18 +00:00
Code Issues Packages Projects Releases Wiki Activity

added text norm for other decoding scripts

Browse Source
This commit is contained in:
JinZr 2024-10-12 16:08:07 +08:00
parent 5492a6a5e2
commit cd96f635c3
3 changed files with 15 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
egs/libritts/ASR/zipformer/ctc_decode.py
View File

@ -122,7 +122,7 @@ import torch
import torch.nn as nn import torch.nn as nn
from asr_datamodule import LibriTTSAsrDataModule from asr_datamodule import LibriTTSAsrDataModule
from lhotse import set_caching_enabled from lhotse import set_caching_enabled
from train import add_model_arguments, get_model, get_params from train import add_model_arguments, get_model, get_params, normalize_text
from icefall.checkpoint import ( from icefall.checkpoint import (
average_checkpoints, average_checkpoints,
@ -949,13 +949,13 @@ def main():
# we need cut ids to display recognition results. # we need cut ids to display recognition results.
args.return_cuts = True args.return_cuts = True
librispeech = LibriTTSAsrDataModule(args) libritts = LibriTTSAsrDataModule(args)
test_clean_cuts = librispeech.test_clean_cuts() test_clean_cuts = libritts.test_clean_cuts().map(normalize_text)
test_other_cuts = librispeech.test_other_cuts() test_other_cuts = libritts.test_other_cuts().map(normalize_text)
test_clean_dl = librispeech.test_dataloaders(test_clean_cuts) test_clean_dl = libritts.test_dataloaders(test_clean_cuts)
test_other_dl = librispeech.test_dataloaders(test_other_cuts) test_other_dl = libritts.test_dataloaders(test_other_cuts)
test_sets = ["test-clean", "test-other"] test_sets = ["test-clean", "test-other"]
test_dl = [test_clean_dl, test_other_dl] test_dl = [test_clean_dl, test_other_dl]

11
egs/libritts/ASR/zipformer/onnx_decode.py
View File

@ -80,6 +80,7 @@ import torch.nn as nn
from asr_datamodule import LibriTTSAsrDataModule from asr_datamodule import LibriTTSAsrDataModule
from k2 import SymbolTable from k2 import SymbolTable
from onnx_pretrained import OnnxModel, greedy_search from onnx_pretrained import OnnxModel, greedy_search
from train import normalize_text
from icefall.utils import setup_logger, store_transcripts, write_error_stats from icefall.utils import setup_logger, store_transcripts, write_error_stats
@ -290,13 +291,13 @@ def main():
# we need cut ids to display recognition results. # we need cut ids to display recognition results.
args.return_cuts = True args.return_cuts = True
librispeech = LibriTTSAsrDataModule(args) libritts = LibriTTSAsrDataModule(args)
test_clean_cuts = librispeech.test_clean_cuts() test_clean_cuts = libritts.test_clean_cuts().map(normalize_text)
test_other_cuts = librispeech.test_other_cuts() test_other_cuts = libritts.test_other_cuts().map(normalize_text)
test_clean_dl = librispeech.test_dataloaders(test_clean_cuts) test_clean_dl = libritts.test_dataloaders(test_clean_cuts)
test_other_dl = librispeech.test_dataloaders(test_other_cuts) test_other_dl = libritts.test_dataloaders(test_other_cuts)
test_sets = ["test-clean", "test-other"] test_sets = ["test-clean", "test-other"]
test_dl = [test_clean_dl, test_other_dl] test_dl = [test_clean_dl, test_other_dl]

6
egs/libritts/ASR/zipformer/streaming_decode.py
View File

@ -52,7 +52,7 @@ from streaming_beam_search import (
) )
from torch import Tensor, nn from torch import Tensor, nn
from torch.nn.utils.rnn import pad_sequence from torch.nn.utils.rnn import pad_sequence
from train import add_model_arguments, get_model, get_params from train import add_model_arguments, get_model, get_params, normalize_text
from icefall.checkpoint import ( from icefall.checkpoint import (
average_checkpoints, average_checkpoints,
@ -866,8 +866,8 @@ def main():
libritts = LibriTTSAsrDataModule(args) libritts = LibriTTSAsrDataModule(args)
test_clean_cuts = libritts.test_clean_cuts() test_clean_cuts = libritts.test_clean_cuts().map(normalize_text)
test_other_cuts = libritts.test_other_cuts() test_other_cuts = libritts.test_other_cuts().map(normalize_text)
test_sets = ["test-clean", "test-other"] test_sets = ["test-clean", "test-other"]
test_cuts = [test_clean_cuts, test_other_cuts] test_cuts = [test_clean_cuts, test_other_cuts]