From 4adb76a6edbca93aae7caa83382e764d7b058f07 Mon Sep 17 00:00:00 2001
From: zhifu gao <zhifu.gzf@alibaba-inc.com>
Date: 星期三, 08 五月 2024 19:21:58 +0800
Subject: [PATCH] Dev gzf exp (#1707)
---
funasr/models/sense_voice/model.py | 30 ++++++++++++++++++++++--------
funasr/models/sense_voice/search.py | 9 ++++-----
funasr/auto/auto_model.py | 20 ++++++++++++--------
3 files changed, 38 insertions(+), 21 deletions(-)
diff --git a/funasr/auto/auto_model.py b/funasr/auto/auto_model.py
index 577c328..97eb325 100644
--- a/funasr/auto/auto_model.py
+++ b/funasr/auto/auto_model.py
@@ -364,7 +364,6 @@
if len(sorted_data) > 0 and len(sorted_data[0]) > 0:
batch_size = max(batch_size, sorted_data[0][0][1] - sorted_data[0][0][0])
- batch_size_ms_cum = 0
beg_idx = 0
beg_asr_total = time.time()
time_speech_total_per_sample = speech_lengths / 16000
@@ -373,19 +372,22 @@
# pbar_sample = tqdm(colour="blue", total=n, dynamic_ncols=True)
all_segments = []
+ max_len_in_batch = 0
+ end_idx = 1
for j, _ in enumerate(range(0, n)):
# pbar_sample.update(1)
- batch_size_ms_cum += sorted_data[j][0][1] - sorted_data[j][0][0]
+ sample_length = sorted_data[j][0][1] - sorted_data[j][0][0]
+ potential_batch_length = max(max_len_in_batch, sample_length) * (j + 1 - beg_idx)
+ # batch_size_ms_cum += sorted_data[j][0][1] - sorted_data[j][0][0]
if (
j < n - 1
- and (batch_size_ms_cum + sorted_data[j + 1][0][1] - sorted_data[j + 1][0][0])
- < batch_size
- and (sorted_data[j + 1][0][1] - sorted_data[j + 1][0][0])
- < batch_size_threshold_ms
+ and sample_length < batch_size_threshold_ms
+ and potential_batch_length < batch_size
):
+ max_len_in_batch = max(max_len_in_batch, sample_length)
+ end_idx += 1
continue
- batch_size_ms_cum = 0
- end_idx = j + 1
+
speech_j, speech_lengths_j = slice_padding_audio_samples(
speech, speech_lengths, sorted_data[beg_idx:end_idx]
)
@@ -410,6 +412,8 @@
)
results[_b]["spk_embedding"] = spk_res[0]["spk_embedding"]
beg_idx = end_idx
+ end_idx += 1
+ max_len_in_batch = sample_length
if len(results) < 1:
continue
results_sorted.extend(results)
diff --git a/funasr/models/sense_voice/model.py b/funasr/models/sense_voice/model.py
index 00bc85b..56e61e7 100644
--- a/funasr/models/sense_voice/model.py
+++ b/funasr/models/sense_voice/model.py
@@ -516,16 +516,23 @@
# Paramterts for rich decoding
self.beam_search.emo_unk = tokenizer.encode(
- DecodingOptions.get("emo_unk_token", "<|SPECIAL_TOKEN_1|>"), allowed_special="all")[0]
+ DecodingOptions.get("emo_unk_token", "<|SPECIAL_TOKEN_1|>"), allowed_special="all"
+ )[0]
self.beam_search.emo_unk_score = 1
self.beam_search.emo_tokens = tokenizer.encode(
- DecodingOptions.get("emo_target_tokens", "<|HAPPY|><|SAD|><|ANGRY|>"), allowed_special="all")
+ DecodingOptions.get("emo_target_tokens", "<|HAPPY|><|SAD|><|ANGRY|>"),
+ allowed_special="all",
+ )
self.beam_search.emo_scores = DecodingOptions.get("emo_target_threshold", [0.1, 0.1, 0.1])
self.beam_search.event_bg_token = tokenizer.encode(
- DecodingOptions.get("gain_tokens_bg", "<|Speech|><|BGM|><|Applause|><|Laughter|>"), allowed_special="all")
+ DecodingOptions.get("gain_tokens_bg", "<|Speech|><|BGM|><|Applause|><|Laughter|>"),
+ allowed_special="all",
+ )
self.beam_search.event_ed_token = tokenizer.encode(
- DecodingOptions.get("gain_tokens_ed", "<|/Speech|><|/BGM|><|/Applause|><|/Laughter|>"), allowed_special="all")
+ DecodingOptions.get("gain_tokens_ed", "<|/Speech|><|/BGM|><|/Applause|><|/Laughter|>"),
+ allowed_special="all",
+ )
self.beam_search.event_score_ga = DecodingOptions.get("gain_tokens_score", [1, 1, 1, 1])
encoder_out, encoder_out_lens = self.encode(
@@ -859,16 +866,23 @@
# Paramterts for rich decoding
self.beam_search.emo_unk = tokenizer.encode(
- DecodingOptions.get("emo_unk_token", "<|SPECIAL_TOKEN_1|>"), allowed_special="all")[0]
+ DecodingOptions.get("emo_unk_token", "<|SPECIAL_TOKEN_1|>"), allowed_special="all"
+ )[0]
self.beam_search.emo_unk_score = 1
self.beam_search.emo_tokens = tokenizer.encode(
- DecodingOptions.get("emo_target_tokens", "<|HAPPY|><|SAD|><|ANGRY|>"), allowed_special="all")
+ DecodingOptions.get("emo_target_tokens", "<|HAPPY|><|SAD|><|ANGRY|>"),
+ allowed_special="all",
+ )
self.beam_search.emo_scores = DecodingOptions.get("emo_target_threshold", [0.1, 0.1, 0.1])
self.beam_search.event_bg_token = tokenizer.encode(
- DecodingOptions.get("gain_tokens_bg", "<|Speech|><|BGM|><|Applause|><|Laughter|>"), allowed_special="all")
+ DecodingOptions.get("gain_tokens_bg", "<|Speech|><|BGM|><|Applause|><|Laughter|>"),
+ allowed_special="all",
+ )
self.beam_search.event_ed_token = tokenizer.encode(
- DecodingOptions.get("gain_tokens_ed", "<|/Speech|><|/BGM|><|/Applause|><|/Laughter|>"), allowed_special="all")
+ DecodingOptions.get("gain_tokens_ed", "<|/Speech|><|/BGM|><|/Applause|><|/Laughter|>"),
+ allowed_special="all",
+ )
self.beam_search.event_score_ga = DecodingOptions.get("gain_tokens_score", [1, 1, 1, 1])
encoder_out, encoder_out_lens = self.encode(
diff --git a/funasr/models/sense_voice/search.py b/funasr/models/sense_voice/search.py
index 4400ce7..3a1a049 100644
--- a/funasr/models/sense_voice/search.py
+++ b/funasr/models/sense_voice/search.py
@@ -54,7 +54,6 @@
event_bg_token: List[int] = field(default_factory=lambda: [58946, 58948, 58950, 58952]),
event_ed_token: List[int] = field(default_factory=lambda: [58947, 58949, 58951, 58953]),
event_score_ga: List[float] = field(default_factory=lambda: [1, 1, 5, 25]),
-
token_list: List[str] = None,
pre_beam_ratio: float = 1.5,
pre_beam_score_key: str = None,
@@ -209,15 +208,16 @@
last_token = yseq[-1]
if last_token in self.emo_tokens + [self.emo_unk]:
- # prevent output event after emotation token
+ # prevent output event after emotation token
score[self.event_bg_token] = -np.inf
- for eve_bg, eve_ed, eve_ga in zip(self.event_bg_token, self.event_ed_token, self.event_score_ga):
+ for eve_bg, eve_ed, eve_ga in zip(
+ self.event_bg_token, self.event_ed_token, self.event_score_ga
+ ):
score_offset = get_score(yseq, eve_bg, eve_ed)
score[eve_bg] += score_offset[0]
score[eve_ed] += score_offset[1]
score[eve_bg] += math.log(eve_ga)
-
score[self.emo_unk] += math.log(self.emo_unk_score)
for emo, emo_th in zip(self.emo_tokens, self.emo_scores):
@@ -231,7 +231,6 @@
scores[k] = struct_score(hyp.yseq, scores[k])
return scores, states
-
def score_partial(
self, hyp: Hypothesis, ids: torch.Tensor, x: torch.Tensor
--
Gitblit v1.9.1