From 8c6d1642f5fbf1d55edb324e35e9fa6e89da25a1 Mon Sep 17 00:00:00 2001
From: Xuning Tan <me@xuning.eu.org>
Date: 星期三, 01 十月 2025 14:45:45 +0800
Subject: [PATCH] fix: correct the deepspeed_config path reference in the finetuning script (#2680)
---
funasr/models/sense_voice/model.py | 65 ++++++++++++++++++++++----------
1 files changed, 44 insertions(+), 21 deletions(-)
diff --git a/funasr/models/sense_voice/model.py b/funasr/models/sense_voice/model.py
index 81feea9..6a29181 100644
--- a/funasr/models/sense_voice/model.py
+++ b/funasr/models/sense_voice/model.py
@@ -19,7 +19,7 @@
from funasr.models.paraformer.search import Hypothesis
-from funasr.models.sense_voice.utils.ctc_alignment import ctc_forced_align
+from .utils.ctc_alignment import ctc_forced_align
class SinusoidalPositionEncoder(torch.nn.Module):
@@ -557,7 +557,7 @@
):
"""Embed positions in tensor."""
maxlen = xs_pad.shape[1]
- masks = sequence_mask(ilens, maxlen = maxlen, device=ilens.device)[:, None, :]
+ masks = sequence_mask(ilens, maxlen=maxlen, device=ilens.device)[:, None, :]
xs_pad *= self.output_size() ** 0.5
@@ -916,32 +916,44 @@
if output_timestamp:
from itertools import groupby
+
timestamp = []
tokens = tokenizer.text2tokens(text)[4:]
- logits_speech = self.ctc.softmax(encoder_out)[i, 4:encoder_out_lens[i].item(), :]
+ token_back_to_id = tokenizer.tokens2ids(tokens)
+ token_ids = []
+ for tok_ls in token_back_to_id:
+ if tok_ls: token_ids.extend(tok_ls)
+ else: token_ids.append(124)
+
+ if len(token_ids) == 0:
+ result_i = {"key": key[i], "text": text}
+ results.append(result_i)
+ continue
+
+ logits_speech = self.ctc.softmax(encoder_out)[i, 4 : encoder_out_lens[i].item(), :]
pred = logits_speech.argmax(-1).cpu()
- logits_speech[pred==self.blank_id, self.blank_id] = 0
+ logits_speech[pred == self.blank_id, self.blank_id] = 0
align = ctc_forced_align(
logits_speech.unsqueeze(0).float(),
- torch.Tensor(token_int[4:]).unsqueeze(0).long().to(logits_speech.device),
- (encoder_out_lens-4).long(),
- torch.tensor(len(token_int)-4).unsqueeze(0).long().to(logits_speech.device),
+ torch.Tensor(token_ids).unsqueeze(0).long().to(logits_speech.device),
+ (encoder_out_lens[i] - 4).long(),
+ torch.tensor(len(token_ids)).unsqueeze(0).long().to(logits_speech.device),
ignore_id=self.ignore_id,
)
- pred = groupby(align[0, :encoder_out_lens[0]])
+ pred = groupby(align[0, : encoder_out_lens[i]])
_start = 0
token_id = 0
ts_max = encoder_out_lens[i] - 4
for pred_token, pred_frame in pred:
_end = _start + len(list(pred_frame))
if pred_token != 0:
- ts_left = max((_start*60-30)/1000, 0)
- ts_right = min((_end*60-30)/1000, (ts_max*60-30)/1000)
+ ts_left = max((_start * 60 - 30) / 1000, 0)
+ ts_right = min((_end * 60 - 30) / 1000, (ts_max * 60 - 30) / 1000)
timestamp.append([tokens[token_id], ts_left, ts_right])
token_id += 1
_start = _end
- timestamp = self.post(timestamp)
- result_i = {"key": key[i], "text": text, "timestamp": timestamp}
+ timestamp, words = self.post(timestamp)
+ result_i = {"key": key[i], "text": text, "timestamp": timestamp, "words": words}
results.append(result_i)
else:
result_i = {"key": key[i], "text": text}
@@ -950,23 +962,35 @@
def post(self, timestamp):
timestamp_new = []
+ words_new = []
+ prev_word = None
for i, t in enumerate(timestamp):
word, start, end = t
- if word == '鈻�':
+ start = int(start * 1000)
+ end = int(end * 1000)
+ if word == "鈻�":
continue
if i == 0:
# timestamp_new.append([word, start, end])
- timestamp_new.append([int(start*1000), int(end*1000)])
- elif word.startswith("鈻�") or len(word) == 1 or not word[1].isalpha():
+ timestamp_new.append([start, end])
+ words_new.append(word)
+ elif word.startswith("鈻�"):
word = word[1:]
- # timestamp_new.append([word, start, end])
- timestamp_new.append([int(start*1000), int(end*1000)])
+ timestamp_new.append([start, end])
+ words_new.append(word)
+ elif prev_word is not None and prev_word.isalpha() and prev_word.isascii() and word.isalpha() and word.isascii():
+ word = prev_word + word
+ timestamp_new[-1][1] = end
+ words_new[-1] = word
else:
# timestamp_new[-1][0] += word
- timestamp_new[-1][1] = int(end*1000)
- return timestamp_new
+ timestamp_new.append([start, end])
+ words_new.append(word)
+ prev_word = word
+ return timestamp_new, words_new
+
def export(self, **kwargs):
- from export_meta import export_rebuild_model
+ from .export_meta import export_rebuild_model
if "max_seq_len" not in kwargs:
kwargs["max_seq_len"] = 512
@@ -974,4 +998,3 @@
return models
return results, meta_data
-
--
Gitblit v1.9.1