Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.
Python asraudiochineseemotion-recognitionfunasrmcp-servermultilingual-asropenai-compatible-apiparaformerpunctuationpytorchreal-time-asrspeaker-diarizationspeech-recognitionspeech-to-textstreaming-asrtranscriptionvllmvoice-activity-detectionwhisper-alternative
update utils
维石02667efa7528
diff --git a/runtime/python/libtorch/funasr_torch/utils/utils.py b/runtime/python/libtorch/funasr_torch/utils/utils.pyindex f85d4e932..ee438521c 100644--- a/runtime/python/libtorch/funasr_torch/utils/utils.py+++ b/runtime/python/libtorch/funasr_torch/utils/utils.py@@ -1,21 +1,25 @@# -*- encoding: utf-8 -*---import functools+import yamlimport logging-import pickle+import functools+import numpy as npfrom pathlib import Pathfrom typing import Any, Dict, Iterable, List, NamedTuple, Set, Tuple, Union-import numpy as np-import yaml---import warnings-root_dir = Path(__file__).resolve().parent-logger_initialized = {}+def pad_list(xs, pad_value, max_len=None):+ n_batch = len(xs)+ if max_len is None:+ max_len = max(x.size(0) for x in xs)+ # pad = xs[0].new(n_batch, max_len, *xs[0].size()[1:]).fill_(pad_value)+ # numpy format+ pad = (np.zeros((n_batch, max_len)) + pad_value).astype(np.int32)+ for i in range(n_batch):+ pad[i, : xs[i].shape[0]] = xs[i]++ return padclass TokenIDConverter:def __init__(