fix typo (#1324)

docs/source/decoding-with-langugage-models/LODR.rst

@@ -56,7 +56,7 @@ during decoding for transducer model:
     \lambda_1 \log p_{\text{Target LM}}\left(y_u|\mathit{x},y_{1:u-1}\right) -
     \lambda_2 \log p_{\text{bi-gram}}\left(y_u|\mathit{x},y_{1:u-1}\right)
 
-In LODR, an additional bi-gram LM estimated on the source domain (e.g training corpus) is required. Comared to DR,
+In LODR, an additional bi-gram LM estimated on the source domain (e.g training corpus) is required. Compared to DR,
 the only difference lies in the choice of source domain LM. According to the original `paper <https://arxiv.org/abs/2203.16776>`_,
 LODR achieves similar performance compared DR in both intra-domain and cross-domain settings.
 As a bi-gram is much faster to evaluate, LODR is usually much faster.

docs/source/model-export/export-ncnn-conv-emformer.rst

@@ -125,7 +125,7 @@ Python code. We have also set up ``PATH`` so that you can use
 .. caution::
 
   Please don't use `<https://github.com/tencent/ncnn>`_.
-  We have made some modifications to the offical `ncnn`_.
+  We have made some modifications to the official `ncnn`_.
 
   We will synchronize `<https://github.com/csukuangfj/ncnn>`_ periodically
   with the official one.

egs/wenetspeech/ASR/pruned_transducer_stateless2/decode.py

@@ -203,7 +203,7 @@ def get_parser():
         "--beam-size",
         type=int,
         default=4,
-        help="""An interger indicating how many candidates we will keep for each
+        help="""An integer indicating how many candidates we will keep for each
         frame. Used only when --decoding-method is beam_search or
         modified_beam_search.""",
     )

egs/wenetspeech/ASR/pruned_transducer_stateless2/finetune.py

@@ -78,7 +78,7 @@ def add_finetune_arguments(parser: argparse.ArgumentParser):
         default=None,
         help="""
         Modules to be initialized. It matches all parameters starting with
-        a specific key. The keys are given with Comma seperated. If None,
+        a specific key. The keys are given with Comma separated. If None,
         all modules will be initialised. For example, if you only want to
         initialise all parameters staring with "encoder", use "encoder";
         if you want to initialise parameters starting with encoder or decoder,

icefall/utils.py

@@ -1977,7 +1977,7 @@ def parse_timestamps_and_texts(
         A k2.Fsa with best_paths.arcs.num_axes() == 3, i.e.
         containing multiple FSAs, which is expected to be the result
         of k2.shortest_path (otherwise the returned values won't
-        be meaningful). Attribtute `labels` is the prediction unit,
+        be meaningful). Attribute `labels` is the prediction unit,
         e.g., phone or BPE tokens. Attribute `aux_labels` is the word index.
       word_table:
         The word symbol table.