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

fix debug

Browse Source
This commit is contained in:
Yuekai Zhang 2024-06-07 09:53:40 +08:00
parent 8bbd06112a
commit 68b99f456f
1 changed files with 4 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
egs/speech_llm/ASR_LLM/whisper_llm_zh/decode.py
View File

@ -290,11 +290,9 @@ def decode_one_batch(
)
generated_ids = model.decode(feature, input_ids.to(device, dtype=torch.long), attention_mask.to(device))
hyps = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
hyps = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)[0]
# hyps = remove_punctuation(hyps)
# hyps = to_simple(hyps)
# hyps = [params.normalizer.normalize(hyp) for hyp in hyps]
hyps = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)
hyps = tokenizer.batch_decode(generated_ids, skip_special_tokens=False)
print(hyps)
texts = batch["supervisions"]["text"]
for i, text in enumerate(texts):
@ -381,7 +379,7 @@ def decode_dataset(
for lm_scale, hyps in hyps_dict.items():
this_batch = []
# assert len(hyps) == len(texts)
assert len(hyps) == len(texts)
for cut_id, hyp_words, ref_text in zip(cut_ids, hyps, texts):
ref_text = normalize_text_alimeeting(ref_text)
ref_words = ref_text.split()