Skip to content

Migrate obsoleted editdistance to rapidfuzz #3105

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
excosy wants to merge 2 commits into
base: main
Choose a base branch
from
Open

Migrate obsoleted editdistance to rapidfuzz #3105

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
dad70eb
Migrate obsoleted editdistance to rapidfuzz
excosy Jul 2, 2026
71af8eb
Adopt gemini suggestions
excosy Jul 2, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 7 additions & 9 deletions funasr/metrics/common.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,8 +9,9 @@
import json
import logging
import sys

from itertools import groupby

from rapidfuzz.distance import Levenshtein
import numpy as np
import six

Expand Down Expand Up @@ -155,7 +156,6 @@ def calculate_cer_ctc(self, ys_hat, ys_pad):
:return: average sentence-level CER score
:rtype float
"""
import editdistance

cers, char_ref_lens = [], []
for i, y in enumerate(ys_hat):
Expand All @@ -175,7 +175,7 @@ def calculate_cer_ctc(self, ys_hat, ys_pad):
hyp_chars = "".join(seq_hat)
ref_chars = "".join(seq_true)
if len(ref_chars) > 0:
cers.append(editdistance.eval(hyp_chars, ref_chars))
cers.append(Levenshtein.distance(hyp_chars, ref_chars))
char_ref_lens.append(len(ref_chars))

cer_ctc = float(sum(cers)) / sum(char_ref_lens) if cers else None
Expand Down Expand Up @@ -214,16 +214,15 @@ def calculate_cer(self, seqs_hat, seqs_true):
:return: average sentence-level CER score
:rtype float
"""
import editdistance

char_eds, char_ref_lens = [], []
for i, seq_hat_text in enumerate(seqs_hat):
seq_true_text = seqs_true[i]
hyp_chars = seq_hat_text.replace(" ", "")
ref_chars = seq_true_text.replace(" ", "")
char_eds.append(editdistance.eval(hyp_chars, ref_chars))
char_eds.append(Levenshtein.distance(hyp_chars, ref_chars))
char_ref_lens.append(len(ref_chars))
return float(sum(char_eds)) / sum(char_ref_lens)
return float(sum(char_eds)) / sum(char_ref_lens) if char_eds else None

def calculate_wer(self, seqs_hat, seqs_true):
"""Calculate sentence-level WER score.
Expand All @@ -233,13 +232,12 @@ def calculate_wer(self, seqs_hat, seqs_true):
:return: average sentence-level WER score
:rtype float
"""
import editdistance

word_eds, word_ref_lens = [], []
for i, seq_hat_text in enumerate(seqs_hat):
seq_true_text = seqs_true[i]
hyp_words = seq_hat_text.split()
ref_words = seq_true_text.split()
word_eds.append(editdistance.eval(hyp_words, ref_words))
word_eds.append(Levenshtein.distance(hyp_words, ref_words))
word_ref_lens.append(len(ref_words))
return float(sum(word_eds)) / sum(word_ref_lens)
return float(sum(word_eds)) / sum(word_ref_lens) if word_eds else None
4 changes: 2 additions & 2 deletions setup.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,7 +35,7 @@
"jaconv",
# Speaker & evaluation
"umap_learn",
"editdistance>=0.5.2",
"rapidfuzz>=3.0.0",
# Optional (training/enhancement)
"torch_complex",
"tensorboardX",
Expand All @@ -44,7 +44,7 @@
],
# train: The modules invoked when training only.
"train": [
"editdistance",
"rapidfuzz>=3.0.0",
],
# all: The modules should be optionally installled due to some reason.
# Please consider moving them to "install" occasionally
Expand Down