Tiếp theoHướng dẫn tiếp theo
Độ chính xác OCR: Tỷ lệ lỗi ký tự và từ
kỹ thuật
HƯỚNG DẪN AI âm thanh
Word error rate, or WER, compares an automatic transcript with a reference by counting word substitutions, deletions and insertions relative to the reference word count.
Nghe hướng dẫn nàyAI thuật lại · ~3 min
It is a standard speech-recognition accuracy measure, but it does not say whether an error changed the meaning or whether a system is fair across speakers. Transcript normalization and sample selection can materially change the number.
Automatic speech recognition turns audio into written words. Word error rate measures how much an output differs from a human reference after aligning the two word sequences. A substitution replaces one reference word, a deletion omits one and an insertion adds an extra word. Add those counts and divide by the number of words in the reference. NIST’s speech-evaluation tools and documentation use these error categories. The denominator is not the number of predicted words, so insertions can make WER exceed 100 percent on a short reference. Suppose a reference says “send five boxes” and the system says “send nine boxes now.” One substitution changes five to nine; one insertion adds now. With three reference words, the WER is two divided by three. Yet that fraction cannot tell whether the wrong number was safety-critical, whether the extra word was harmless or whether a different paraphrase was useful. WER treats words as edit units, not their consequences. Evaluation choices matter. The reference transcript may itself be uncertain in noisy speech. Consistent rules are needed for contractions, punctuation, capitalization, numbers, disfluencies and spelling. A system output “twenty one” may be judged differently against “21” depending on normalization. A single overall WER can also conceal worse performance for particular accents, languages, age groups, microphones or background noise. Report slices with enough examples and inspect where errors occur, not only their aggregate count. For a product, pair WER with task outcomes and human review where words carry special meaning. Names, medication amounts or negations may deserve explicit checks. Real-time systems also need latency metrics; a low final WER does not mean the words appeared promptly or remained stable as the person spoke. Keep reference data independent of model tuning and describe the normalization procedure so the score can be reproduced.
A transcription team counts one substitution when “fifteen” is recognized as “fifty.”
An evaluator measures WER separately for noisy calls and clean studio speech rather than hiding the difference.
A medical transcription reviewer checks clinically important names even when overall WER is low.
Two labs agree on punctuation and number normalization before comparing ASR systems.
Speech systems will improve across accents and noisy settings, but WER will remain useful because it is simple and comparable under a fixed protocol. Better evaluations can add named-entity, numeric and negation errors, speaker-level slices and human task outcomes. Streaming products should report how long partial and final words take, alongside WER. Datasets need careful reference transcripts and disclosed normalization rules. A single aggregate score should not erase a severe error on a high-impact term or a subgroup. The most informative reports will combine WER with the specific consequences of mistakes in the application.
Bản tin miễn phí
Ba câu chuyện AI đã được xác minh vào mỗi buổi sáng các ngày trong tuần, được viết bằng tiếng Anh đơn giản. Miễn phí mãi mãi, không có quảng cáo.
Một email mỗi ngày trong tuần. Hủy đăng ký trong một cú nhấp chuột. Chúng tôi không bao giờ bán hoặc chia sẻ địa chỉ của bạn.
Tự kiểm tra
Phản hồi tức thì về mọi câu trả lời và chứng chỉ có thể chia sẻ với ID có thể xác minh sau khi bạn vượt qua khóa học.
Hỗ trợ giáo dục AI miễn phí. AI Understanding là tổ chức phi lợi nhuận 501(c)(3) — không bao giờ có quảng cáo, không có tường phí. Hãy quyên góp
Word error rate, or WER, compares an automatic transcript with a reference by counting word substitutions, deletions and insertions relative to the reference word count. It is a standard speech-recognition accuracy measure, but it does not say whether an error changed the meaning or whether a system is fair across speakers. Transcript normalization and sample selection can materially change the number.
A transcription team counts one substitution when “fifteen” is recognized as “fifty.” An evaluator measures WER separately for noisy calls and clean studio speech rather than hiding the difference. A medical transcription reviewer checks clinically important names even when overall WER is low. Two labs agree on punctuation and number normalization before comparing ASR systems.
Speech systems will improve across accents and noisy settings, but WER will remain useful because it is simple and comparable under a fixed protocol. Better evaluations can add named-entity, numeric and negation errors, speaker-level slices and human task outcomes. Streaming products should report how long partial and final words take, alongside WER. Datasets need careful reference transcripts and disclosed normalization rules. A single aggregate score should not erase a severe error on a high-impact term or a subgroup. The most informative reports will combine WER with the specific consequences of mistakes in the application.
Tiếp tục học hỏi
Đã chọn thêm hướng dẫn cho chủ đề này
Tiếp theoHướng dẫn tiếp theo
Độ chính xác OCR: Tỷ lệ lỗi ký tự và từ
kỹ thuật