The multimodal corpus of Tibetan teaching uses video and audio documents as the corpus, which truly reflects the teaching activities and contents; furthermore, it has great potential in promoting Tibetan teaching and researches. Compared with the textual corpus, the multimodal corpus of Tibetan teaching has the advantage of the authenticity and the timeliness of the corpus. This article mainly discusses the construction of multimodal corpus, the corpus collection, the segmentation of annotation, the tools of database construction, and the retrieval and replacement, etc. This article analyzes the application value of multimodal corpus in Tibetan teaching and puts forward the applied countermeasures in teaching domain.

As a special phenomenon of human society, language is always accompanied with human beings. And it has always been in a dynamic and open ecosystem. It gives us a lot of inspiration. We must observe and discuss the interrelation and mutual influence between language and its surrounding environment as well as social reality while studying language. In view of this, this paper adopts the method of comparative analysis, and focuses on three aspects from the perspective of sociolinguistics, which include language learning and national identity as well as individual identity, language vision and language planning, the promotion of Mandarin and the protection of endangered languages. And finally it comes to the conclusion. Language has the dual functions of national identity and individual identity. There is no fundamental conflict between the promotion of mandarin and the protection of minority languages in China. China's language policy and language planning should be based on China's national conditions. We should not only consider China's needs of current economic, political, and cultural development, but also combine its goals of long-term development and the individual future development.

No abstract

Abstract. Taking mandarin voiced consonant as its study objects, it has used PAS6600 to collect 3 male and female speakers' articulation signals including speech signal, airflow signal and air-pressure signal in order to explore the mechanism of aerodynamic for mandarin voiced consonant. The study found that: 1）in the process of pronunciation [m] and [n] both in peak expiratory airflow（PEA） and expiratory volume （EV）is less than the [l] and [r], the peak air pressure（PAP）of [m] and [n] is also less than [l] and [n], namely in the process of pronunciation, peak expiratory airflow, expiratory volume, peak air pressure of mandarin voiced consonant nasals is less than the lateral sound and approximant. 2）Mandarin voiced consonant peak air pressure, peak expiratory airflow and expiratory volume of men are more than women, this is due to generally male lung capacity is greater than female. Apart from Expiratory airflow duration (EAD) nasal [n], there is on difference between men and women. Mandarin Chinese voiced consonant airflow pressure study further describes a voiced consonant pronunciation mechanism, to further distinguish the nasal, lateral sound and approximant.

No abstract