diff --git a/docs/source/recipes/aishell/conformer_ctc.rst b/docs/source/recipes/aishell/conformer_ctc.rst
index 2dcf0c728..75a2a8eca 100644
--- a/docs/source/recipes/aishell/conformer_ctc.rst
+++ b/docs/source/recipes/aishell/conformer_ctc.rst
@@ -1,4 +1,4 @@
-Confromer CTC
+Conformer CTC
 =============
 
 This tutorial shows you how to run a conformer ctc model
diff --git a/egs/aishell/ASR/local/compile_hlg.py b/egs/aishell/ASR/local/compile_hlg.py
deleted file mode 100755
index 098d5d6a3..000000000
--- a/egs/aishell/ASR/local/compile_hlg.py
+++ /dev/null
@@ -1,156 +0,0 @@
-#!/usr/bin/env python3
-# Copyright    2021  Xiaomi Corp.        (authors: Fangjun Kuang)
-#
-# See ../../../../LICENSE for clarification regarding multiple authors
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-
-
-"""
-This script takes as input lang_dir and generates HLG from
-
-    - H, the ctc topology, built from tokens contained in lang_dir/lexicon.txt
-    - L, the lexicon, built from lang_dir/L_disambig.pt
-
-        Caution: We use a lexicon that contains disambiguation symbols
-
-    - G, the LM, built from data/lm/G_3_gram.fst.txt
-
-The generated HLG is saved in $lang_dir/HLG.pt
-"""
-import argparse
-import logging
-from pathlib import Path
-
-import k2
-import torch
-
-from icefall.lexicon import Lexicon
-
-
-def get_args():
-    parser = argparse.ArgumentParser()
-    parser.add_argument(
-        "--lang-dir",
-        type=str,
-        help="""Input and output directory.
-        """,
-    )
-
-    return parser.parse_args()
-
-
-def compile_HLG(lang_dir: str) -> k2.Fsa:
-    """
-    Args:
-      lang_dir:
-        The language directory, e.g., data/lang_phone or data/lang_bpe_5000.
-
-    Return:
-      An FSA representing HLG.
-    """
-    lexicon = Lexicon(lang_dir)
-    max_token_id = max(lexicon.tokens)
-    logging.info(f"Building ctc_topo. max_token_id: {max_token_id}")
-    H = k2.ctc_topo(max_token_id)
-    L = k2.Fsa.from_dict(torch.load(f"{lang_dir}/L_disambig.pt"))
-
-    if Path("data/lm/G_3_gram.pt").is_file():
-        logging.info("Loading pre-compiled G_3_gram")
-        d = torch.load("data/lm/G_3_gram.pt")
-        G = k2.Fsa.from_dict(d)
-    else:
-        logging.info("Loading G_3_gram.fst.txt")
-        with open("data/lm/G_3_gram.fst.txt") as f:
-            G = k2.Fsa.from_openfst(f.read(), acceptor=False)
-            torch.save(G.as_dict(), "data/lm/G_3_gram.pt")
-
-    first_token_disambig_id = lexicon.token_table["#0"]
-    first_word_disambig_id = lexicon.word_table["#0"]
-
-    L = k2.arc_sort(L)
-    G = k2.arc_sort(G)
-
-    logging.info("Intersecting L and G")
-    LG = k2.compose(L, G)
-    logging.info(f"LG shape: {LG.shape}")
-
-    logging.info("Connecting LG")
-    LG = k2.connect(LG)
-    logging.info(f"LG shape after k2.connect: {LG.shape}")
-
-    logging.info(type(LG.aux_labels))
-    logging.info("Determinizing LG")
-
-    LG = k2.determinize(LG)
-    logging.info(type(LG.aux_labels))
-
-    logging.info("Connecting LG after k2.determinize")
-    LG = k2.connect(LG)
-
-    logging.info("Removing disambiguation symbols on LG")
-
-    LG.labels[LG.labels >= first_token_disambig_id] = 0
-
-    assert isinstance(LG.aux_labels, k2.RaggedTensor)
-    LG.aux_labels.values[LG.aux_labels.values >= first_word_disambig_id] = 0
-
-    LG = k2.remove_epsilon(LG)
-    logging.info(f"LG shape after k2.remove_epsilon: {LG.shape}")
-
-    LG = k2.connect(LG)
-    LG.aux_labels = LG.aux_labels.remove_values_eq(0)
-
-    logging.info("Arc sorting LG")
-    LG = k2.arc_sort(LG)
-
-    logging.info("Composing H and LG")
-    # CAUTION: The name of the inner_labels is fixed
-    # to `tokens`. If you want to change it, please
-    # also change other places in icefall that are using
-    # it.
-    HLG = k2.compose(H, LG, inner_labels="tokens")
-
-    logging.info("Connecting LG")
-    HLG = k2.connect(HLG)
-
-    logging.info("Arc sorting LG")
-    HLG = k2.arc_sort(HLG)
-    logging.info(f"HLG.shape: {HLG.shape}")
-
-    return HLG
-
-
-def main():
-    args = get_args()
-    lang_dir = Path(args.lang_dir)
-
-    if (lang_dir / "HLG.pt").is_file():
-        logging.info(f"{lang_dir}/HLG.pt already exists - skipping")
-        return
-
-    logging.info(f"Processing {lang_dir}")
-
-    HLG = compile_HLG(lang_dir)
-    logging.info(f"Saving HLG.pt to {lang_dir}")
-    torch.save(HLG.as_dict(), f"{lang_dir}/HLG.pt")
-
-
-if __name__ == "__main__":
-    formatter = (
-        "%(asctime)s %(levelname)s [%(filename)s:%(lineno)d] %(message)s"
-    )
-
-    logging.basicConfig(format=formatter, level=logging.INFO)
-
-    main()
diff --git a/egs/aishell/ASR/local/compile_hlg.py b/egs/aishell/ASR/local/compile_hlg.py
new file mode 120000
index 000000000..471aa7fb4
--- /dev/null
+++ b/egs/aishell/ASR/local/compile_hlg.py
@@ -0,0 +1 @@
+../../../librispeech/ASR/local/compile_hlg.py
\ No newline at end of file
diff --git a/egs/aishell/ASR/local/compute_fbank_musan.py b/egs/aishell/ASR/local/compute_fbank_musan.py
deleted file mode 100755
index e79bdafb1..000000000
--- a/egs/aishell/ASR/local/compute_fbank_musan.py
+++ /dev/null
@@ -1,110 +0,0 @@
-#!/usr/bin/env python3
-# Copyright    2021  Xiaomi Corp.        (authors: Fangjun Kuang)
-#
-# See ../../../../LICENSE for clarification regarding multiple authors
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-
-
-"""
-This file computes fbank features of the musan dataset.
-It looks for manifests in the directory data/manifests.
-
-The generated fbank features are saved in data/fbank.
-"""
-
-import argparse
-import logging
-import os
-from pathlib import Path
-
-import torch
-from lhotse import CutSet, Fbank, FbankConfig, LilcomHdf5Writer, combine
-from lhotse.recipes.utils import read_manifests_if_cached
-
-from icefall.utils import get_executor
-
-# Torch's multithreaded behavior needs to be disabled or
-# it wastes a lot of CPU and slow things down.
-# Do this outside of main() in case it needs to take effect
-# even when we are not invoking the main (e.g. when spawning subprocesses).
-torch.set_num_threads(1)
-torch.set_num_interop_threads(1)
-
-
-def compute_fbank_musan(num_mel_bins: int = 80):
-    src_dir = Path("data/manifests")
-    output_dir = Path("data/fbank")
-    num_jobs = min(15, os.cpu_count())
-
-    dataset_parts = (
-        "music",
-        "speech",
-        "noise",
-    )
-    manifests = read_manifests_if_cached(
-        dataset_parts=dataset_parts, output_dir=src_dir
-    )
-    assert manifests is not None
-
-    musan_cuts_path = output_dir / "cuts_musan.json.gz"
-
-    if musan_cuts_path.is_file():
-        logging.info(f"{musan_cuts_path} already exists - skipping")
-        return
-
-    logging.info("Extracting features for Musan")
-
-    extractor = Fbank(FbankConfig(num_mel_bins=num_mel_bins))
-
-    with get_executor() as ex:  # Initialize the executor only once.
-        # create chunks of Musan with duration 5 - 10 seconds
-        musan_cuts = (
-            CutSet.from_manifests(
-                recordings=combine(
-                    part["recordings"] for part in manifests.values()
-                )
-            )
-            .cut_into_windows(10.0)
-            .filter(lambda c: c.duration > 5)
-            .compute_and_store_features(
-                extractor=extractor,
-                storage_path=f"{output_dir}/feats_musan",
-                num_jobs=num_jobs if ex is None else 80,
-                executor=ex,
-                storage_type=LilcomHdf5Writer,
-            )
-        )
-        musan_cuts.to_json(musan_cuts_path)
-
-
-def get_args():
-    parser = argparse.ArgumentParser()
-    parser.add_argument(
-        "--num-mel-bins",
-        type=int,
-        default=80,
-        help="""The number of mel bins for Fbank""",
-    )
-
-    return parser.parse_args()
-
-
-if __name__ == "__main__":
-    formatter = (
-        "%(asctime)s %(levelname)s [%(filename)s:%(lineno)d] %(message)s"
-    )
-
-    logging.basicConfig(format=formatter, level=logging.INFO)
-    args = get_args()
-    compute_fbank_musan(num_mel_bins=args.num_mel_bins)
diff --git a/egs/aishell/ASR/local/compute_fbank_musan.py b/egs/aishell/ASR/local/compute_fbank_musan.py
new file mode 120000
index 000000000..5833f2484
--- /dev/null
+++ b/egs/aishell/ASR/local/compute_fbank_musan.py
@@ -0,0 +1 @@
+../../../librispeech/ASR/local/compute_fbank_musan.py
\ No newline at end of file
diff --git a/egs/aishell/ASR/local/convert_transcript_words_to_tokens.py b/egs/aishell/ASR/local/convert_transcript_words_to_tokens.py
deleted file mode 100755
index 133499c8b..000000000
--- a/egs/aishell/ASR/local/convert_transcript_words_to_tokens.py
+++ /dev/null
@@ -1,107 +0,0 @@
-#!/usr/bin/env python3
-
-# Copyright 2021 Xiaomi Corporation (Author: Fangjun Kuang)
-"""
-Convert a transcript file containing words to a corpus file containing tokens
-for LM training with the help of a lexicon.
-
-If the lexicon contains phones, the resulting LM will be a phone LM; If the
-lexicon contains word pieces, the resulting LM will be a word piece LM.
-
-If a word has multiple pronunciations, the one that appears first in the lexicon
-is kept; others are removed.
-
-If the input transcript is:
-
-    hello zoo world hello
-    world zoo
-    foo zoo world hellO
-
-and if the lexicon is
-
-    <UNK> SPN
-    hello h e l l o 2
-    hello h e l l o
-    world w o r l d
-    zoo z o o
-
-Then the output is
-
-    h e l l o 2 z o o w o r l d h e l l o 2
-    w o r l d z o o
-    SPN z o o w o r l d SPN
-"""
-
-import argparse
-from pathlib import Path
-from typing import Dict, List
-
-from generate_unique_lexicon import filter_multiple_pronunications
-
-from icefall.lexicon import read_lexicon
-
-
-def get_args():
-    parser = argparse.ArgumentParser()
-    parser.add_argument(
-        "--transcript",
-        type=str,
-        help="The input transcript file."
-        "We assume that the transcript file consists of "
-        "lines. Each line consists of space separated words.",
-    )
-    parser.add_argument("--lexicon", type=str, help="The input lexicon file.")
-    parser.add_argument(
-        "--oov", type=str, default="<UNK>", help="The OOV word."
-    )
-
-    return parser.parse_args()
-
-
-def process_line(
-    lexicon: Dict[str, List[str]], line: str, oov_token: str
-) -> None:
-    """
-    Args:
-      lexicon:
-        A dict containing pronunciations. Its keys are words and values
-        are pronunciations (i.e., tokens).
-      line:
-        A line of transcript consisting of space(s) separated words.
-      oov_token:
-        The pronunciation of the oov word if a word in `line` is not present
-        in the lexicon.
-    Returns:
-      Return None.
-    """
-    s = ""
-    words = line.strip().split()
-    for i, w in enumerate(words):
-        tokens = lexicon.get(w, oov_token)
-        s += " ".join(tokens)
-        s += " "
-    print(s.strip())
-
-
-def main():
-    args = get_args()
-    assert Path(args.lexicon).is_file()
-    assert Path(args.transcript).is_file()
-    assert len(args.oov) > 0
-
-    # Only the first pronunciation of a word is kept
-    lexicon = filter_multiple_pronunications(read_lexicon(args.lexicon))
-
-    lexicon = dict(lexicon)
-
-    assert args.oov in lexicon
-
-    oov_token = lexicon[args.oov]
-
-    with open(args.transcript) as f:
-        for line in f:
-            process_line(lexicon=lexicon, line=line, oov_token=oov_token)
-
-
-if __name__ == "__main__":
-    main()
diff --git a/egs/aishell/ASR/local/convert_transcript_words_to_tokens.py b/egs/aishell/ASR/local/convert_transcript_words_to_tokens.py
new file mode 120000
index 000000000..2ce13fd69
--- /dev/null
+++ b/egs/aishell/ASR/local/convert_transcript_words_to_tokens.py
@@ -0,0 +1 @@
+../../../librispeech/ASR/local/convert_transcript_words_to_tokens.py
\ No newline at end of file
diff --git a/egs/aishell/ASR/local/generate_unique_lexicon.py b/egs/aishell/ASR/local/generate_unique_lexicon.py
deleted file mode 100755
index 566c0743d..000000000
--- a/egs/aishell/ASR/local/generate_unique_lexicon.py
+++ /dev/null
@@ -1,100 +0,0 @@
-#!/usr/bin/env python3
-# Copyright    2021  Xiaomi Corp.        (authors: Fangjun Kuang)
-#
-# See ../../../../LICENSE for clarification regarding multiple authors
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-
-"""
-This file takes as input a lexicon.txt and output a new lexicon,
-in which each word has a unique pronunciation.
-
-The way to do this is to keep only the first pronunciation of a word
-in lexicon.txt.
-"""
-
-
-import argparse
-import logging
-from pathlib import Path
-from typing import List, Tuple
-
-from icefall.lexicon import read_lexicon, write_lexicon
-
-
-def get_args():
-    parser = argparse.ArgumentParser()
-    parser.add_argument(
-        "--lang-dir",
-        type=str,
-        help="""Input and output directory.
-        It should contain a file lexicon.txt.
-        This file will generate a new file uniq_lexicon.txt
-        in it.
-        """,
-    )
-
-    return parser.parse_args()
-
-
-def filter_multiple_pronunications(
-    lexicon: List[Tuple[str, List[str]]]
-) -> List[Tuple[str, List[str]]]:
-    """Remove multiple pronunciations of words from a lexicon.
-
-    If a word has more than one pronunciation in the lexicon, only
-    the first one is kept, while other pronunciations are removed
-    from the lexicon.
-
-    Args:
-      lexicon:
-        The input lexicon, containing a list of (word, [p1, p2, ..., pn]),
-        where "p1, p2, ..., pn" are the pronunciations of the "word".
-    Returns:
-      Return a new lexicon where each word has a unique pronunciation.
-    """
-    seen = set()
-    ans = []
-
-    for word, tokens in lexicon:
-        if word in seen:
-            continue
-        seen.add(word)
-        ans.append((word, tokens))
-    return ans
-
-
-def main():
-    args = get_args()
-    lang_dir = Path(args.lang_dir)
-
-    lexicon_filename = lang_dir / "lexicon.txt"
-
-    in_lexicon = read_lexicon(lexicon_filename)
-
-    out_lexicon = filter_multiple_pronunications(in_lexicon)
-
-    write_lexicon(lang_dir / "uniq_lexicon.txt", out_lexicon)
-
-    logging.info(f"Number of entries in lexicon.txt: {len(in_lexicon)}")
-    logging.info(f"Number of entries in uniq_lexicon.txt: {len(out_lexicon)}")
-
-
-if __name__ == "__main__":
-    formatter = (
-        "%(asctime)s %(levelname)s [%(filename)s:%(lineno)d] %(message)s"
-    )
-
-    logging.basicConfig(format=formatter, level=logging.INFO)
-
-    main()
diff --git a/egs/aishell/ASR/local/generate_unique_lexicon.py b/egs/aishell/ASR/local/generate_unique_lexicon.py
new file mode 120000
index 000000000..c0aea1403
--- /dev/null
+++ b/egs/aishell/ASR/local/generate_unique_lexicon.py
@@ -0,0 +1 @@
+../../../librispeech/ASR/local/generate_unique_lexicon.py
\ No newline at end of file
diff --git a/egs/timit/ASR/local/compute_fbank_musan.py b/egs/timit/ASR/local/compute_fbank_musan.py
deleted file mode 100644
index d44524e70..000000000
--- a/egs/timit/ASR/local/compute_fbank_musan.py
+++ /dev/null
@@ -1,97 +0,0 @@
-#!/usr/bin/env python3
-# Copyright    2021  Xiaomi Corp.        (authors: Fangjun Kuang)
-#
-# See ../../../../LICENSE for clarification regarding multiple authors
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-
-
-"""
-This file computes fbank features of the musan dataset.
-It looks for manifests in the directory data/manifests.
-
-The generated fbank features are saved in data/fbank.
-"""
-
-import logging
-import os
-from pathlib import Path
-
-import torch
-from lhotse import CutSet, Fbank, FbankConfig, LilcomHdf5Writer, combine
-from lhotse.recipes.utils import read_manifests_if_cached
-
-from icefall.utils import get_executor
-
-# Torch's multithreaded behavior needs to be disabled or
-# it wastes a lot of CPU and slow things down.
-# Do this outside of main() in case it needs to take effect
-# even when we are not invoking the main (e.g. when spawning subprocesses).
-torch.set_num_threads(1)
-torch.set_num_interop_threads(1)
-
-
-def compute_fbank_musan():
-    src_dir = Path("data/manifests")
-    output_dir = Path("data/fbank")
-    num_jobs = min(15, os.cpu_count())
-    num_mel_bins = 80
-
-    dataset_parts = (
-        "music",
-        "speech",
-        "noise",
-    )
-    manifests = read_manifests_if_cached(
-        dataset_parts=dataset_parts, output_dir=src_dir
-    )
-    assert manifests is not None
-
-    musan_cuts_path = output_dir / "cuts_musan.json.gz"
-
-    if musan_cuts_path.is_file():
-        logging.info(f"{musan_cuts_path} already exists - skipping")
-        return
-
-    logging.info("Extracting features for Musan")
-
-    extractor = Fbank(FbankConfig(num_mel_bins=num_mel_bins))
-
-    with get_executor() as ex:  # Initialize the executor only once.
-        # create chunks of Musan with duration 5 - 10 seconds
-        musan_cuts = (
-            CutSet.from_manifests(
-                recordings=combine(
-                    part["recordings"] for part in manifests.values()
-                )
-            )
-            .cut_into_windows(10.0)
-            .filter(lambda c: c.duration > 5)
-            .compute_and_store_features(
-                extractor=extractor,
-                storage_path=f"{output_dir}/feats_musan",
-                num_jobs=num_jobs if ex is None else 80,
-                executor=ex,
-                storage_type=LilcomHdf5Writer,
-            )
-        )
-        musan_cuts.to_json(musan_cuts_path)
-
-
-if __name__ == "__main__":
-    formatter = (
-        "%(asctime)s %(levelname)s [%(filename)s:%(lineno)d] %(message)s"
-    )
-
-    logging.basicConfig(format=formatter, level=logging.INFO)
-    compute_fbank_musan()
diff --git a/egs/timit/ASR/local/compute_fbank_musan.py b/egs/timit/ASR/local/compute_fbank_musan.py
new file mode 120000
index 000000000..5833f2484
--- /dev/null
+++ b/egs/timit/ASR/local/compute_fbank_musan.py
@@ -0,0 +1 @@
+../../../librispeech/ASR/local/compute_fbank_musan.py
\ No newline at end of file
diff --git a/egs/timit/ASR/shared b/egs/timit/ASR/shared
deleted file mode 100644
index 4c5e91438..000000000
--- a/egs/timit/ASR/shared
+++ /dev/null
@@ -1 +0,0 @@
-../../../icefall/shared/
\ No newline at end of file
diff --git a/egs/timit/ASR/shared b/egs/timit/ASR/shared
new file mode 120000
index 000000000..4cbd91a7e
--- /dev/null
+++ b/egs/timit/ASR/shared
@@ -0,0 +1 @@
+../../../icefall/shared
\ No newline at end of file