Acupuncture combined with speech rehabilitation training for post-stroke spasmodic dysphonia:a multicenter randomized controlled trial

Objective To observe the clinical efficacy of acupuncture combined with speech rehabilitation training for post-stroke spasmodic dysphonia and compare the differences in efficacy among the therapy of acupuncture combined with speech rehabilitation training and the monotherapy.Methods Two hundred and seventy cases of patients with post-stroke spasmodic dysarthria were randomly divided into three groups with the ratio of 1：1：1.Group A：acupuncture combined with speech rehabilitation training group（90 cases）,group B：acupuncture group（90 cases）,group C：speech rehabilitation training group（90 cases）.In the group B,acupuncture treatment was given at Baihui（百会 GV 20）,JTnjin（金津 EX-HN 12） and Yuye（玉液 EX-HN13） as well as tongue-three needles.In the group C,the treatment of speech rehabilitation training was provided.The two treatments mentioned above were combined in the group A.Patients were treated once a day for a month with improved Frenchay dysarthria rating scale as the indices of therapeutic effect evaluation.Results ① Group A：the cured and markedly effective rate was 88.7%,and total effective rate was 94.3%;group B：the cured and markedly effective rate was 44.2%and total effective rate was 81.4%;group C：the cured and markedly effective rate was 23.5%and total effective rate was 61.2%.Both the cured and markedly effective rate and the total effective rate in the group A were significantly higher than those in the group B or group C（all P0.05）;both the cured and markedly effective rate and the total effective rate in the group B were higher than those of group C（both P0.05）;② In comparison of functional recovery of tongue in accordance with the Frenchay dysarthria rating scale,the recovery rate of the tongue-stationary state was 71.74%in the group A,18.87%in the group B and 4.44%in the group C;the recovery rate of tongue lolling out was 66.23%in the group A,27.63%in the group B and 1.59%in the group C;the recovery rate of tongue up and down motion was 44.19%in the group A,4.94%in the group B and 1.35%in the group C;the recovery rate of lateral motion was 40.24%in the group A,7.59%in the group B and 0.00%in the group C;the recovery rate of alternating motion was 29.07%in the group A,7.14%in the group B and 1.23%in the group C;the recovery rate of speech was 29.07%in the group A,5.88%in the group B and 1.22%in the group C.In the three groups,the recovery rates of stationary state and tongue lolling out were superior to those of up and down movement,lateral movement,alternating movement and speech（all P0.05）.Conclusion The clinical efficacy of acupuncture combined with speech rehabilitation training for patients with post-stroke spasmodic dysarthria is significant,and the efficacy of acupuncture is superior to that of speech rehabilitation training;as for functional recovery of tongue like stationary state and tongue out,the therapy of acupuncture combined with speech rehabilitation training is effective.

Past review,current progress,and challenges ahead on the cocktail party problem

The cocktail party problem,i.e.,tracing and recognizing the speech of a specific speaker when multiple speakers talk simultaneously,is one of the critical problems yet to be solved to enable the wide application of automatic speech recognition(ASR) systems.In this overview paper,we review the techniques proposed in the last two decades in attacking this problem.We focus our discussions on the speech separation problem given its central role in the cocktail party environment,and describe the conventional single-channel techniques such as computational auditory scene analysis(CASA),non-negative matrix factorization(NMF) and generative models,the conventional multi-channel techniques such as beamforming and multi-channel blind source separation,and the newly developed deep learning-based techniques,such as deep clustering(DPCL),the deep attractor network(DANet),and permutation invariant training(PIT).We also present techniques developed to improve ASR accuracy and speaker identification in the cocktail party environment.We argue effectively exploiting information in the microphone array,the acoustic training set,and the language itself using a more powerful model.Better optimization ob jective and techniques will be the approach to solving the cocktail party problem.