speech_recognition/FunASR
1
0
Fork 0
mirror of https://github.com/modelscope/FunASR synced 2025-09-15 14:48:36 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

fix bug

Browse Source
This commit is contained in:
游雁 2024-06-08 21:08:54 +08:00
parent 2191795f74
commit 9fd3c47fd7
1 changed files with 3 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
funasr/models/llm_asr/model.py
View File

@ -19,6 +19,7 @@ from funasr.utils import postprocess_utils
from funasr.utils.datadir_writer import DatadirWriter from funasr.utils.datadir_writer import DatadirWriter
from funasr.register import tables from funasr.register import tables
from funasr.train_utils.device_funcs import to_device from funasr.train_utils.device_funcs import to_device
import traceback
@tables.register("model_classes", "LLMASR") @tables.register("model_classes", "LLMASR")
@ -489,6 +490,7 @@ class LLMASR2(nn.Module):
fbank_fake_len = fbank_fake_lens[batch_idx].item() fbank_fake_len = fbank_fake_lens[batch_idx].item()
fbank_beg_idx = fbank_beg[batch_idx, 0].item() fbank_beg_idx = fbank_beg[batch_idx, 0].item()
min_len = min(fbank_fake_len, inputs_embeds.shape[1] - fbank_beg_idx) min_len = min(fbank_fake_len, inputs_embeds.shape[1] - fbank_beg_idx)
try: try:
inputs_embeds[batch_idx, fbank_beg_idx : fbank_beg_idx + min_len, :] = encoder_out[ inputs_embeds[batch_idx, fbank_beg_idx : fbank_beg_idx + min_len, :] = encoder_out[
batch_idx, :min_len, : batch_idx, :min_len, :
@ -499,7 +501,7 @@ class LLMASR2(nn.Module):
f"batch_idx: {batch_idx}, inputs_embeds: {inputs_embeds.shape}, fbank_beg_idx: {fbank_beg_idx}, min_len: {min_len}, fbank_fake_len: {fbank_fake_len}" f"batch_idx: {batch_idx}, inputs_embeds: {inputs_embeds.shape}, fbank_beg_idx: {fbank_beg_idx}, min_len: {min_len}, fbank_fake_len: {fbank_fake_len}"
) )
fbank_fake_len = encoder_out_lens[batch_idx].item() fbank_fake_len = encoder_out_lens[batch_idx].item()
min_len = min(fbank_fake_len, inputs_embeds.shape[1] - fbank_beg_idx) min_len = min(fbank_fake_len, min_len)
inputs_embeds[batch_idx, fbank_beg_idx : fbank_beg_idx + min_len, :] = encoder_out[ inputs_embeds[batch_idx, fbank_beg_idx : fbank_beg_idx + min_len, :] = encoder_out[
batch_idx, :min_len, : batch_idx, :min_len, :
] ]