From 4ace5a95b052d338947fc88809a440ccd55cf6b4 Mon Sep 17 00:00:00 2001
From: 游雁 <zhifu.gzf@alibaba-inc.com>
Date: 星期四, 16 十一月 2023 16:39:52 +0800
Subject: [PATCH] funasr pages
---
funasr/datasets/large_datasets/dataset.py | 71 ++++++++++++++++++++++++++++++-----
1 files changed, 60 insertions(+), 11 deletions(-)
diff --git a/funasr/datasets/large_datasets/dataset.py b/funasr/datasets/large_datasets/dataset.py
index 33ed13a..adfe4f6 100644
--- a/funasr/datasets/large_datasets/dataset.py
+++ b/funasr/datasets/large_datasets/dataset.py
@@ -6,6 +6,8 @@
import torch
import torch.distributed as dist
import torchaudio
+import numpy as np
+import soundfile
from kaldiio import ReadHelper
from torch.utils.data import IterableDataset
@@ -105,8 +107,8 @@
if data_type == "kaldi_ark":
ark_reader = ReadHelper('ark:{}'.format(data_file))
reader_list.append(ark_reader)
- elif data_type == "text" or data_type == "sound":
- text_reader = open(data_file, "r")
+ elif data_type == "text" or data_type == "sound" or data_type == 'text_hotword':
+ text_reader = open(data_file, "r", encoding="utf-8")
reader_list.append(text_reader)
elif data_type == "none":
continue
@@ -123,7 +125,14 @@
sample_dict["key"] = key
elif data_type == "sound":
key, path = item.strip().split()
- waveform, sampling_rate = torchaudio.load(path)
+ try:
+ waveform, sampling_rate = torchaudio.load(path)
+ except:
+ waveform, sampling_rate = soundfile.read(path, dtype='float32')
+ if waveform.ndim == 2:
+ waveform = waveform[:, 0]
+ waveform = np.expand_dims(waveform, axis=0)
+ waveform = torch.tensor(waveform)
if self.frontend_conf is not None:
if sampling_rate != self.frontend_conf["fs"]:
waveform = torchaudio.transforms.Resample(orig_freq=sampling_rate,
@@ -135,11 +144,25 @@
speed = random.choice(self.speed_perturb)
if speed != 1.0:
mat, _ = torchaudio.sox_effects.apply_effects_tensor(
- mat, sampling_rate, [['speed', str(speed)], ['rate', str(sampling_rate)]])
+ torch.tensor(mat).view(1, -1), sampling_rate, [['speed', str(speed)], ['rate', str(sampling_rate)]])
+ mat = mat.view(-1).numpy()
sample_dict[data_name] = mat
sample_dict["sampling_rate"] = sampling_rate
if data_name == "speech":
sample_dict["key"] = key
+ elif data_type == "text_hotword":
+ text = item
+ segs = text.strip().split()
+ sample_dict[data_name] = segs[1:]
+ if "key" not in sample_dict:
+ sample_dict["key"] = segs[0]
+ sample_dict['hw_tag'] = 1
+ elif data_type == "text_nospace":
+ text = item
+ segs = text.strip().split(maxsplit=1)
+ sample_dict[data_name] = [x for x in segs[1]]
+ if "key" not in sample_dict:
+ sample_dict["key"] = segs[0]
else:
text = item
segs = text.strip().split()
@@ -177,17 +200,43 @@
shuffle = conf.get('shuffle', True)
data_names = conf.get("data_names", "speech,text")
data_types = conf.get("data_types", "kaldi_ark,text")
- dataset = AudioDataset(scp_lists, data_names, data_types, frontend_conf=frontend_conf, shuffle=shuffle,
- speed_perturb=speed_perturb, mode=mode)
+
+ pre_hwfile = conf.get("pre_hwlist", None)
+ # pre_prob = conf.get("pre_prob", 0) # unused yet
+ if pre_hwfile is not None:
+ pre_hwlist = []
+ with open(pre_hwfile, 'r', encoding="utf-8") as fin:
+ for line in fin.readlines():
+ pre_hwlist.append(line.strip())
+ else:
+ pre_hwlist = None
+
+ hw_config = {"sample_rate": conf.get("sample_rate", 0.6),
+ "double_rate": conf.get("double_rate", 0.1),
+ "hotword_min_length": conf.get("hotword_min_length", 2),
+ "hotword_max_length": conf.get("hotword_max_length", 8),
+ "pre_prob": conf.get("pre_prob", 0.0),
+ "pre_hwlist": pre_hwlist}
+
+
+
+ dataset = AudioDataset(scp_lists,
+ data_names,
+ data_types,
+ frontend_conf=frontend_conf,
+ shuffle=shuffle,
+ speed_perturb=speed_perturb,
+ mode=mode,
+ )
+
+ if "text" in data_names:
+ vocab = {'vocab': dict, 'seg_dict': seg_dict, 'punc_dict': punc_dict, 'bpe_tokenizer': bpe_tokenizer, 'hw_config': hw_config}
+ tokenize_fn = partial(tokenize, **vocab)
+ dataset = MapperIterDataPipe(dataset, fn=tokenize_fn)
filter_conf = conf.get('filter_conf', {})
filter_fn = partial(filter, **filter_conf)
dataset = FilterIterDataPipe(dataset, fn=filter_fn)
-
- if "text" in data_names:
- vocab = {'vocab': dict, 'seg_dict': seg_dict, 'punc_dict': punc_dict, 'bpe_tokenizer': bpe_tokenizer}
- tokenize_fn = partial(tokenize, **vocab)
- dataset = MapperIterDataPipe(dataset, fn=tokenize_fn)
if shuffle:
buffer_conf = conf.get('shuffle_conf', {})
--
Gitblit v1.9.1