mirrors/icefall
1
0
Fork 0
mirror of https://github.com/k2-fsa/icefall.git synced 2025-09-03 06:04:18 +00:00
Code Issues Packages Projects Releases Wiki Activity

Minor fixes.

Browse Source
This commit is contained in:
Fangjun Kuang 2022-03-22 14:53:17 +08:00
parent 4187ad0d06
commit 4808dd3fdb
1 changed files with 42 additions and 22 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

64
egs/librispeech/ASR/pruned_transducer_stateless/pretrained.py
View File

@ -50,7 +50,12 @@ import kaldifeat
import sentencepiece as spm import sentencepiece as spm
import torch import torch
import torchaudio import torchaudio
from beam_search import beam_search, greedy_search, modified_beam_search from beam_search import (
beam_search,
greedy_search,
greedy_search_batch,
modified_beam_search,
)
from torch.nn.utils.rnn import pad_sequence from torch.nn.utils.rnn import pad_sequence
from train import get_params, get_transducer_model from train import get_params, get_transducer_model
@ -224,28 +229,43 @@ def main():
if params.method == "beam_search": if params.method == "beam_search":
msg += f" with beam size {params.beam_size}" msg += f" with beam size {params.beam_size}"
logging.info(msg) logging.info(msg)
for i in range(num_waves): if params.method == "modified_beam_search":
# fmt: off hyp_tokens = modified_beam_search(
encoder_out_i = encoder_out[i:i+1, :encoder_out_lens[i]] model=model,
# fmt: on encoder_out=encoder_out,
if params.method == "greedy_search": beam=params.beam_size,
hyp = greedy_search( )
model=model,
encoder_out=encoder_out_i,
max_sym_per_frame=params.max_sym_per_frame,
)
elif params.method == "beam_search":
hyp = beam_search(
model=model, encoder_out=encoder_out_i, beam=params.beam_size
)
elif params.method == "modified_beam_search":
hyp = modified_beam_search(
model=model, encoder_out=encoder_out_i, beam=params.beam_size
)
else:
raise ValueError(f"Unsupported method: {params.method}")
hyps.append(sp.decode(hyp).split()) for hyp in sp.decode(hyp_tokens):
hyps.append(hyp.split())
elif params.method == "greedy_search" and params.max_sym_per_frame == 1:
hyp_tokens = greedy_search_batch(
model=model,
encoder_out=encoder_out,
)
for hyp in sp.decode(hyp_tokens):
hyps.append(hyp.split())
else:
for i in range(num_waves):
# fmt: off
encoder_out_i = encoder_out[i:i+1, :encoder_out_lens[i]]
# fmt: on
if params.method == "greedy_search":
hyp = greedy_search(
model=model,
encoder_out=encoder_out_i,
max_sym_per_frame=params.max_sym_per_frame,
)
elif params.method == "beam_search":
hyp = beam_search(
model=model,
encoder_out=encoder_out_i,
beam=params.beam_size,
)
else:
raise ValueError(f"Unsupported method: {params.method}")
hyps.append(sp.decode(hyp).split())
s = "\n" s = "\n"
for filename, hyp in zip(params.sound_files, hyps): for filename, hyp in zip(params.sound_files, hyps):