From 0ac06c029edb57e2dcacd64da2a05869a2f7364d Mon Sep 17 00:00:00 2001
From: 志浩 <neo.dzh@alibaba-inc.com>
Date: 星期四, 16 三月 2023 19:24:15 +0800
Subject: [PATCH] fixbug path_name_type_list can [[any,str,str],[any,str,str]]
---
funasr/datasets/iterable_dataset.py | 15 +++++++++++++--
1 files changed, 13 insertions(+), 2 deletions(-)
diff --git a/funasr/datasets/iterable_dataset.py b/funasr/datasets/iterable_dataset.py
index 2ac37b2..3798280 100644
--- a/funasr/datasets/iterable_dataset.py
+++ b/funasr/datasets/iterable_dataset.py
@@ -8,6 +8,7 @@
from typing import Iterator
from typing import Tuple
from typing import Union
+from typing import List
import kaldiio
import numpy as np
@@ -20,7 +21,7 @@
from funasr.datasets.dataset import ESPnetDataset
-SUPPORT_AUDIO_TYPE_SETS = ['flac', 'mp3', 'm4a', 'ogg', 'opus', 'wav', 'wma']
+SUPPORT_AUDIO_TYPE_SETS = ['flac', 'mp3', 'ogg', 'opus', 'wav', 'pcm']
def load_kaldi(input):
retval = kaldiio.load_mat(input)
@@ -60,9 +61,14 @@
array = np.frombuffer((middle_data.astype(dtype) - offset) / abs_max, dtype=np.float32)
return array
+def load_pcm(input):
+ with open(input,"rb") as f:
+ bytes = f.read()
+ return load_bytes(bytes)
DATA_TYPES = {
"sound": lambda x: torchaudio.load(x)[0][0].numpy(),
+ "pcm": load_pcm,
"kaldi_ark": load_kaldi,
"bytes": load_bytes,
"waveform": lambda x: x,
@@ -122,7 +128,7 @@
non_iterable_list = []
self.path_name_type_list = []
- if not isinstance(path_name_type_list[0], Tuple):
+ if not isinstance(path_name_type_list[0], (Tuple, List)):
path = path_name_type_list[0]
name = path_name_type_list[1]
_type = path_name_type_list[2]
@@ -219,6 +225,9 @@
if audio_type not in SUPPORT_AUDIO_TYPE_SETS:
raise NotImplementedError(
f'Not supported audio type: {audio_type}')
+ if audio_type == "pcm":
+ _type = "pcm"
+
func = DATA_TYPES[_type]
array = func(value)
if self.fs is not None and name == "speech":
@@ -318,6 +327,8 @@
if audio_type not in SUPPORT_AUDIO_TYPE_SETS:
raise NotImplementedError(
f'Not supported audio type: {audio_type}')
+ if audio_type == "pcm":
+ _type = "pcm"
func = DATA_TYPES[_type]
# Load entry
array = func(value)
--
Gitblit v1.9.1