IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
US-0652535
(2010-01-05)
|
등록번호 |
US-8494857
(2013-07-23)
|
발명자
/ 주소 |
|
출원인 / 주소 |
- Regents of the University of Minnesota
|
대리인 / 주소 |
Shumaker & Sieffert, P.A.
|
인용정보 |
피인용 횟수 :
12 인용 특허 :
198 |
초록
▼
Techniques are described for automatically measuring fluency of a patient's speech based on prosodic characteristics thereof. The prosodic characteristics may include statistics regarding silent pauses, filled pauses, repetitions, or fundamental frequency of the patient's speech. The statistics may
Techniques are described for automatically measuring fluency of a patient's speech based on prosodic characteristics thereof. The prosodic characteristics may include statistics regarding silent pauses, filled pauses, repetitions, or fundamental frequency of the patient's speech. The statistics may include a count, average number of occurrences, duration, average duration, frequency of occurrence, standard deviation, or other statistics. In one embodiment, a method includes receiving an audio sample that includes speech of a patient, analyzing the audio sample to identify prosodic characteristics of the speech of the patient, and automatically measuring fluency of the speech of the patient based on the prosodic characteristics. These techniques may present several advantages, such as objectively measuring fluency of a patient's speech without requiring a manual transcription or other manual intervention in the analysis process.
대표청구항
▼
1. A method comprising: receiving, with a speech analysis device, an audio sample that includes speech of a patient;analyzing, with the speech analysis device, the audio sample to identify phonemes from the speech of the patient;analyzing, with the speech analysis device, the identified phonemes to
1. A method comprising: receiving, with a speech analysis device, an audio sample that includes speech of a patient;analyzing, with the speech analysis device, the audio sample to identify phonemes from the speech of the patient;analyzing, with the speech analysis device, the identified phonemes to identify prosodic characteristics of the speech of the patient; andautomatically measuring, with the speech analysis device, fluency of the speech of the patient based on the prosodic characteristics. 2. The method of claim 1, wherein analyzing the identified phonemes to identify prosodic characteristics comprises identifying pauses between the identified phonemes of the speech of the patient from the audio sample. 3. The method of claim 2, wherein identifying pauses comprises identifying silent pauses between the identified phonemes. 4. The method of claim 2, wherein identifying pauses comprises identifying filled pauses that include at least one filled pause phoneme. 5. The method of claim 2, wherein automatically measuring fluency comprises counting a number of pauses that occur in the speech of the patient of the audio sample. 6. The method of claim 5, wherein counting a number of pauses comprises counting a number of silent pauses and counting a number of filled pauses. 7. The method of claim 2, wherein automatically measuring fluency comprises identifying a frequency of pauses of the speech of the patient of the audio sample. 8. The method of claim 2, wherein automatically measuring fluency comprises identifying an average duration of pauses of the speech of the patient of the audio sample. 9. The method of claim 1, wherein analyzing the identified phonemes to identify prosodic characteristics comprises identifying repetitions of one or more of the identified phonemes. 10. The method of claim 9, wherein automatically measuring fluency comprises counting a number of repetitions that occur in the identified phonemes. 11. The method of claim 9, wherein identifying repetitions comprises: calculating a first spectrogram for a first portion of the audio sample;calculating a second spectrogram for a second portion of the audio sample;comparing the first spectrogram to the second spectrogram to determine similarity between the first spectrogram and the second spectrogram; anddetermining that the second portion is a repetition of the first portion when the similarity between the first spectrogram and the second spectrogram exceeds a threshold. 12. The method of claim 11, wherein comparing the first spectrogram to the second spectrogram comprises: computing a cosine angle between each of a plurality of vectors of the first spectrogram and a corresponding one of a plurality of vectors of the second spectrogram;determining a mean of the computed cosine angles; anddetermining that the similarity between the first spectrogram and the second spectrogram is the computed mean of the determined cosine angles. 13. The method of claim 1, wherein analyzing the identified phonemes to identify prosodic characteristics comprises identifying a fundamental frequency of the identified phonemes. 14. The method of claim 1, further comprising diagnosing a disorder of the patient according to the fluency of the speech of the patient. 15. The method of claim 14, wherein the disorder comprises at least one of fronto-temporal dementia, Alzheimer's disease, schizophrenia, epilepsy, an autism spectrum disorder, and a side effect caused by medication. 16. The method of claim 1, further comprising monitoring changes in the fluency of the speech of the patient over time. 17. The method of claim 16, further comprising modifying a therapy regimen based on changes in the fluency of the speech of the patient. 18. The method of claim 1, further comprising determining efficacy of a drug based on the fluency of the speech of the patient. 19. A non-transitory computer-readable storage medium comprising instructions for causing a programmable processor to: receive an audio sample that includes speech of a patient;analyze the audio sample to identify phonemes from the speech of the patient;analyze the identified phonemes to identify prosodic characteristics of the speech of the patient; andautomatically measure fluency of the speech of the patient based on the prosodic characteristics. 20. A method comprising: providing an audio sample that includes speech of a patient to a speech analysis device, wherein the speech analysis device analyzes the audio sample to identify phonemes from the speech of the patient and to analyze the identified phonemes to identify prosodic characteristics of the speech of the patient and automatically measures fluency of the speech of the patient based on the prosodic characteristics; andreceiving an indication of the fluency of the speech of the patient from the speech analysis device. 21. The method of claim 20, further comprising diagnosing a disorder of the patient according to the fluency of the speech of the patient. 22. The method of claim 21, wherein the disorder comprises at least one of fronto-temporal dementia, Alzheimer's disease, schizophrenia, epilepsy, an autism spectrum disorder, and a side effect caused by medication. 23. The method of claim 20, further comprising monitoring changes in the fluency of the speech of the patient over time. 24. The method of claim 23, further comprising modifying a therapy regimen based on changes in the fluency of the speech of the patient. 25. The method of claim 20, further comprising determining efficacy of a drug based on the fluency of the speech of the patient.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.