Skip to content

measures

measures #

Convenience methods for calculating a number of similarity error measures between a reference and hypothesis sentence. These measures are commonly used to measure the performance for an automatic speech recognition (ASR) system.

The following measures are implemented:

  • Word Error Rate (WER), which is where this library got its name from. This has long been (and arguably still is) the de facto standard for computing ASR performance.
  • Match Error Rate (MER)
  • Word Information Lost (WIL)
  • Word Information Preserved (WIP)
  • Character Error Rate (CER)

Note that these functions merely call jiwer.process_words and jiwer.process_characters. It is more efficient to call process_words or process_characters and access the results from the jiwer.WordOutput and jiwer.CharacterOutput classes.

cer #

cer(
    reference=None,
    hypothesis=None,
    reference_transform=cer_default,
    hypothesis_transform=cer_default,
    return_dict=False,
    truth=None,
    truth_transform=None,
)

Calculate the character error rate (CER) between one or more reference and hypothesis sentences.

Parameters:

Name Type Description Default
reference Union[str, List[str]]

The reference sentence(s)

 None
hypothesis Union[str, List[str]]

The hypothesis sentence(s)

 None
reference_transform Union[Compose, AbstractTransform]

The transformation(s) to apply to the reference string(s)

 cer_default
hypothesis_transform Union[Compose, AbstractTransform]

The transformation(s) to apply to the hypothesis string(s)

 cer_default
return_dict bool

Deprecated option to return the more results in a dict instead of returning only the cer as a single float value

 False
truth Union[str, List[str]]

Deprecated, renamed to reference

 None
truth_transform Union[Compose, AbstractTransform]

Deprecated, renamed to reference_transform

 None
Deprecated

Argument return_dict will be deprecated. Please use jiwer.process_characters instead.

Arguments truth and truth_transform have been renamed to respectively reference and reference_transform. Therefore, the keyword arguments truth and truth_transform will be removed in the next release. At the same time, reference and reference_transform will lose their default value.

Returns:

Type Description
float

The character error rate of the given reference and hypothesis sentence(s).
Source code in jiwer/measures.py 
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
def cer(
    reference: Union[str, List[str]] = None,
    hypothesis: Union[str, List[str]] = None,
    reference_transform: Union[tr.Compose, tr.AbstractTransform] = cer_default,
    hypothesis_transform: Union[tr.Compose, tr.AbstractTransform] = cer_default,
    return_dict: bool = False,
    truth: Union[str, List[str]] = None,
    truth_transform: Union[tr.Compose, tr.AbstractTransform] = None,
) -> Union[float, Dict[str, Any]]:
    """
    Calculate the character error rate (CER) between one or more reference and
    hypothesis sentences.

    Args:
        reference: The reference sentence(s)
        hypothesis: The hypothesis sentence(s)
        reference_transform: The transformation(s) to apply to the reference string(s)
        hypothesis_transform: The transformation(s) to apply to the hypothesis string(s)
        return_dict: Deprecated option to return the more results in a dict instead of
                     returning only the cer as a single float value
        truth: Deprecated, renamed to `reference`
        truth_transform: Deprecated, renamed to `reference_transform`

    Deprecated:
        Argument `return_dict` will be deprecated. Please use
        [jiwer.process_characters][process.process_characters] instead.

        Arguments `truth` and `truth_transform` have been renamed to respectively
        `reference` and `reference_transform`. Therefore, the keyword arguments
         `truth` and `truth_transform` will be removed in the next release.
         At the same time, `reference` and `reference_transform` will lose their
         default value.

    Returns:
        (float): The character error rate of the given reference and hypothesis
                 sentence(s).
    """
    (
        reference,
        hypothesis,
        reference_transform,
        hypothesis_transform,
    ) = _deprecate_truth(
        reference=reference,
        hypothesis=hypothesis,
        truth=truth,
        reference_transform=reference_transform,
        truth_transform=truth_transform,
        hypothesis_transform=hypothesis_transform,
    )

    output = process_characters(
        reference, hypothesis, reference_transform, hypothesis_transform
    )

    if return_dict:
        warnings.warn(
            DeprecationWarning(
                "`return_dict` is deprecated, "
                "please use jiwer.process_characters() instead."
            )
        )
        return {
            "cer": output.cer,
            "hits": output.hits,
            "substitutions": output.substitutions,
            "deletions": output.deletions,
            "insertions": output.insertions,
        }
    else:
        return output.cer

compute_measures #

compute_measures(
    truth,
    hypothesis,
    truth_transform=wer_default,
    hypothesis_transform=wer_default,
)

Efficiently computes all measures using only one function call.

Deprecated

Deprecated method. Superseded by jiwer.process_words. This method will be removed on next release.

Parameters:

Name Type Description Default
truth Union[str, List[str]]

The reference sentence(s)

 required
hypothesis Union[str, List[str]]

The hypothesis sentence(s)

 required
truth_transform Union[Compose, AbstractTransform]

The transformation(s) to apply to the reference string(s)

 wer_default
hypothesis_transform Union[Compose, AbstractTransform]

The transformation(s) to apply to the hypothesis string(s)

 wer_default

Returns:

Type Description
dict

A dictionary containing key-value pairs for all measures.
Source code in jiwer/measures.py 
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
def compute_measures(
    truth: Union[str, List[str]],
    hypothesis: Union[str, List[str]],
    truth_transform: Union[tr.Compose, tr.AbstractTransform] = wer_default,
    hypothesis_transform: Union[tr.Compose, tr.AbstractTransform] = wer_default,
) -> Dict[str, Any]:
    """
    Efficiently computes all measures using only one function call.

    Deprecated:
        Deprecated method. Superseded by [jiwer.process_words][process.process_words].
        This method will be removed on next release.

    Args:
        truth: The reference sentence(s)
        hypothesis: The hypothesis sentence(s)
        truth_transform: The transformation(s) to apply to the reference string(s)
        hypothesis_transform: The transformation(s) to apply to the hypothesis string(s)

    Returns:
        (dict): A dictionary containing key-value pairs for all measures.

    """
    warnings.warn(
        DeprecationWarning(
            "jiwer.compute_measures() is deprecated. Please use jiwer.process_words()."
        )
    )

    output = process_words(
        reference=truth,
        hypothesis=hypothesis,
        reference_transform=truth_transform,
        hypothesis_transform=hypothesis_transform,
    )

    return {
        "wer": output.wer,
        "mer": output.mer,
        "wil": output.wil,
        "wip": output.wip,
        "hits": output.hits,
        "substitutions": output.substitutions,
        "deletions": output.deletions,
        "insertions": output.insertions,
        "ops": output.alignments,
        "truth": output.references,
        "hypothesis": output.hypotheses,
    }

mer #

mer(
    reference=None,
    hypothesis=None,
    reference_transform=wer_default,
    hypothesis_transform=wer_default,
    truth=None,
    truth_transform=None,
)

Calculate the match error rate (MER) between one or more reference and hypothesis sentences.

Parameters:

Name Type Description Default
reference Union[str, List[str]]

The reference sentence(s)

 None
hypothesis Union[str, List[str]]

The hypothesis sentence(s)

 None
reference_transform Union[Compose, AbstractTransform]

The transformation(s) to apply to the reference string(s)

 wer_default
hypothesis_transform Union[Compose, AbstractTransform]

The transformation(s) to apply to the hypothesis string(s)

 wer_default
truth Union[str, List[str]]

Deprecated, renamed to reference

 None
truth_transform Union[Compose, AbstractTransform]

Deprecated, renamed to reference_transform

 None
Deprecated

Arguments truth and truth_transform have been renamed to respectively reference and reference_transform. Therefore, the keyword arguments truth and truth_transform will be removed in the next release. At the same time, reference and reference_transform will lose their default value.

Returns:

Type Description
float

The match error rate of the given reference and hypothesis sentence(s).
Source code in jiwer/measures.py 
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
def mer(
    reference: Union[str, List[str]] = None,
    hypothesis: Union[str, List[str]] = None,
    reference_transform: Union[tr.Compose, tr.AbstractTransform] = wer_default,
    hypothesis_transform: Union[tr.Compose, tr.AbstractTransform] = wer_default,
    truth: Union[str, List[str]] = None,
    truth_transform: Union[tr.Compose, tr.AbstractTransform] = None,
) -> float:
    """
    Calculate the match error rate (MER) between one or more reference and
    hypothesis sentences.

    Args:
        reference: The reference sentence(s)
        hypothesis: The hypothesis sentence(s)
        reference_transform: The transformation(s) to apply to the reference string(s)
        hypothesis_transform: The transformation(s) to apply to the hypothesis string(s)
        truth: Deprecated, renamed to `reference`
        truth_transform: Deprecated, renamed to `reference_transform`

    Deprecated:
        Arguments `truth` and `truth_transform` have been renamed to respectively
        `reference` and `reference_transform`. Therefore, the keyword arguments
         `truth` and `truth_transform` will be removed in the next release.
         At the same time, `reference` and `reference_transform` will lose their
         default value.

    Returns:
        (float): The match error rate of the given reference and
                 hypothesis sentence(s).
    """
    (
        reference,
        hypothesis,
        reference_transform,
        hypothesis_transform,
    ) = _deprecate_truth(
        reference=reference,
        hypothesis=hypothesis,
        truth=truth,
        reference_transform=reference_transform,
        truth_transform=truth_transform,
        hypothesis_transform=hypothesis_transform,
    )

    output = process_words(
        reference, hypothesis, reference_transform, hypothesis_transform
    )

    return output.mer

wer #

wer(
    reference=None,
    hypothesis=None,
    reference_transform=wer_default,
    hypothesis_transform=wer_default,
    truth=None,
    truth_transform=None,
)

Calculate the word error rate (WER) between one or more reference and hypothesis sentences.

Parameters:

Name Type Description Default
reference Union[str, List[str]]

The reference sentence(s)

 None
hypothesis Union[str, List[str]]

The hypothesis sentence(s)

 None
reference_transform Union[Compose, AbstractTransform]

The transformation(s) to apply to the reference string(s)

 wer_default
hypothesis_transform Union[Compose, AbstractTransform]

The transformation(s) to apply to the hypothesis string(s)

 wer_default
truth Union[str, List[str]]

Deprecated, renamed to reference

 None
truth_transform Union[Compose, AbstractTransform]

Deprecated, renamed to reference_transform

 None
Deprecated

Arguments truth and truth_transform have been renamed to respectively reference and reference_transform. Therefore, the keyword arguments truth and truth_transform will be removed in the next release. At the same time, reference and reference_transform will lose their default value.

Returns:

Type Description
float

The word error rate of the given reference and hypothesis sentence(s).
Source code in jiwer/measures.py 
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
def wer(
    reference: Union[str, List[str]] = None,
    hypothesis: Union[str, List[str]] = None,
    reference_transform: Union[tr.Compose, tr.AbstractTransform] = wer_default,
    hypothesis_transform: Union[tr.Compose, tr.AbstractTransform] = wer_default,
    truth: Union[str, List[str]] = None,
    truth_transform: Union[tr.Compose, tr.AbstractTransform] = None,
) -> float:
    """
    Calculate the word error rate (WER) between one or more reference and
    hypothesis sentences.

    Args:
        reference: The reference sentence(s)
        hypothesis: The hypothesis sentence(s)
        reference_transform: The transformation(s) to apply to the reference string(s)
        hypothesis_transform: The transformation(s) to apply to the hypothesis string(s)
        truth: Deprecated, renamed to `reference`
        truth_transform: Deprecated, renamed to `reference_transform`

    Deprecated:
        Arguments `truth` and `truth_transform` have been renamed to respectively
        `reference` and `reference_transform`. Therefore, the keyword arguments
         `truth` and `truth_transform` will be removed in the next release.
         At the same time, `reference` and `reference_transform` will lose their
         default value.

    Returns:
        (float): The word error rate of the given reference and
                 hypothesis sentence(s).
    """
    (
        reference,
        hypothesis,
        reference_transform,
        hypothesis_transform,
    ) = _deprecate_truth(
        reference=reference,
        hypothesis=hypothesis,
        truth=truth,
        reference_transform=reference_transform,
        truth_transform=truth_transform,
        hypothesis_transform=hypothesis_transform,
    )

    output = process_words(
        reference, hypothesis, reference_transform, hypothesis_transform
    )
    return output.wer

wil #

wil(
    reference=None,
    hypothesis=None,
    reference_transform=wer_default,
    hypothesis_transform=wer_default,
    truth=None,
    truth_transform=None,
)

Calculate the word information lost (WIL) between one or more reference and hypothesis sentences.

Parameters:

Name Type Description Default
reference Union[str, List[str]]

The reference sentence(s)

 None
hypothesis Union[str, List[str]]

The hypothesis sentence(s)

 None
reference_transform Union[Compose, AbstractTransform]

The transformation(s) to apply to the reference string(s)

 wer_default
hypothesis_transform Union[Compose, AbstractTransform]

The transformation(s) to apply to the hypothesis string(s)

 wer_default
truth Union[str, List[str]]

Deprecated, renamed to reference

 None
truth_transform Union[Compose, AbstractTransform]

Deprecated, renamed to reference_transform

 None
Deprecated

Arguments truth and truth_transform have been renamed to respectively reference and reference_transform. Therefore, the keyword arguments truth and truth_transform will be removed in the next release. At the same time, reference and reference_transform will lose their default value.

Returns:

Type Description
float

The word information lost of the given reference and hypothesis sentence(s).
Source code in jiwer/measures.py 
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
def wil(
    reference: Union[str, List[str]] = None,
    hypothesis: Union[str, List[str]] = None,
    reference_transform: Union[tr.Compose, tr.AbstractTransform] = wer_default,
    hypothesis_transform: Union[tr.Compose, tr.AbstractTransform] = wer_default,
    truth: Union[str, List[str]] = None,
    truth_transform: Union[tr.Compose, tr.AbstractTransform] = None,
) -> float:
    """
    Calculate the word information lost (WIL) between one or more reference and
    hypothesis sentences.

    Args:
        reference: The reference sentence(s)
        hypothesis: The hypothesis sentence(s)
        reference_transform: The transformation(s) to apply to the reference string(s)
        hypothesis_transform: The transformation(s) to apply to the hypothesis string(s)
        truth: Deprecated, renamed to `reference`
        truth_transform: Deprecated, renamed to `reference_transform`

    Deprecated:
        Arguments `truth` and `truth_transform` have been renamed to respectively
        `reference` and `reference_transform`. Therefore, the keyword arguments
        `truth` and `truth_transform` will be removed in the next release.
         At the same time, `reference` and `reference_transform` will lose their
         default value.

    Returns:
        (float): The word information lost of the given reference and
                 hypothesis sentence(s).
    """
    (
        reference,
        hypothesis,
        reference_transform,
        hypothesis_transform,
    ) = _deprecate_truth(
        reference=reference,
        hypothesis=hypothesis,
        truth=truth,
        reference_transform=reference_transform,
        truth_transform=truth_transform,
        hypothesis_transform=hypothesis_transform,
    )

    output = process_words(
        reference, hypothesis, reference_transform, hypothesis_transform
    )

    return output.wil

wip #

wip(
    reference=None,
    hypothesis=None,
    reference_transform=wer_default,
    hypothesis_transform=wer_default,
    truth=None,
    truth_transform=None,
)

Calculate the word information preserved (WIP) between one or more reference and hypothesis sentences.

Parameters:

Name Type Description Default
reference Union[str, List[str]]

The reference sentence(s)

 None
hypothesis Union[str, List[str]]

The hypothesis sentence(s)

 None
reference_transform Union[Compose, AbstractTransform]

The transformation(s) to apply to the reference string(s)

 wer_default
hypothesis_transform Union[Compose, AbstractTransform]

The transformation(s) to apply to the hypothesis string(s)

 wer_default
truth Union[str, List[str]]

Deprecated, renamed to reference

 None
truth_transform Union[Compose, AbstractTransform]

Deprecated, renamed to reference_transform

 None
Deprecated

Arguments truth and truth_transform have been renamed to respectively reference and reference_transform. Therefore, the keyword arguments truth and truth_transform will be removed in the next release. At the same time, reference and reference_transform will lose their default value.

Returns:

Type Description
float

The word information preserved of the given reference and hypothesis sentence(s).
Source code in jiwer/measures.py 
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
def wip(
    reference: Union[str, List[str]] = None,
    hypothesis: Union[str, List[str]] = None,
    reference_transform: Union[tr.Compose, tr.AbstractTransform] = wer_default,
    hypothesis_transform: Union[tr.Compose, tr.AbstractTransform] = wer_default,
    truth: Union[str, List[str]] = None,
    truth_transform: Union[tr.Compose, tr.AbstractTransform] = None,
) -> float:
    """
    Calculate the word information preserved (WIP) between one or more reference and
    hypothesis sentences.

    Args:
        reference: The reference sentence(s)
        hypothesis: The hypothesis sentence(s)
        reference_transform: The transformation(s) to apply to the reference string(s)
        hypothesis_transform: The transformation(s) to apply to the hypothesis string(s)
        truth: Deprecated, renamed to `reference`
        truth_transform: Deprecated, renamed to `reference_transform`

    Deprecated:
        Arguments `truth` and `truth_transform` have been renamed to respectively
        `reference` and `reference_transform`. Therefore, the keyword arguments
         `truth` and `truth_transform` will be removed in the next release.
         At the same time, `reference` and `reference_transform` will lose their
         default value.

    Returns:
        (float): The word information preserved of the given reference and
                 hypothesis sentence(s).
    """
    (
        reference,
        hypothesis,
        reference_transform,
        hypothesis_transform,
    ) = _deprecate_truth(
        reference=reference,
        hypothesis=hypothesis,
        truth=truth,
        reference_transform=reference_transform,
        truth_transform=truth_transform,
        hypothesis_transform=hypothesis_transform,
    )

    output = process_words(
        reference, hypothesis, reference_transform, hypothesis_transform
    )

    return output.wip