• The Journal of Korea Robotics Society
• /
• v.13 no.1
• /
• pp.16-25
• /
• 2018

The talking head (TH) indicates an utterance face animation generated based on text and voice input. In this paper, we propose the generation method of TH with facial expression and intonation by speech input only. The problem of generating TH from speech can be regarded as a regression problem from the acoustic feature sequence to the facial code sequence which is a low dimensional vector representation that can efficiently encode and decode a face image. This regression was modeled by bidirectional RNN and trained by using SAVEE database of the front utterance face animation database as training data. The proposed method is able to generate TH with facial expression and intonation TH by using acoustic features such as MFCC, dynamic elements of MFCC, energy, and F0. According to the experiments, the configuration of the BLSTM layer of the first and second layers of bidirectional RNN was able to predict the face code best. For the evaluation, a questionnaire survey was conducted for 62 persons who watched TH animations, generated by the proposed method and the previous method. As a result, 77% of the respondents answered that the proposed method generated TH, which matches well with the speech.

• Trans-
• /
• v.7
• /
• pp.103-121
• /
• 2019

In some documentary films, there are interviews, but in some other ones, there aren't any. As for the documentaries with interviews, the director uses talking heads to suggest their own arguments and to have audience agree. It is evident that interviews play a key role since they lead the story of a human-related documentary. Some directors prefer interviews where directors' questions and interviewees' answers are both heard. On the other hand, there a re directors who carry out the story just by using the answers from the interviewees. Then, it becomes crucial that we understand both of these styles and have new perspectives when watching documentary films. Ethics has always been considered one of the most important factors in documentaries and since they are believed to be "true," the influence documentaries have on the society is enormous. However, possibilities of exaggerating or manipulating the "truth" always exist. Therefore, it is important for the audience to identify the intentions of the director. As a matter of fact, there have been countless researches being done, but it is difficult to find studies that discuss the ways interviews are used in documentary films. The two different styles mentioned above do have huge differences in terms of directing methods. Thus, taking a look at differences from diverse angles will help us better understand the essentials of documentary films. This study will take examples from the interviews in Michael Moore's (2002), (2007) and observe how the flow of interviews can change when the voice of the director gets involved. There will also be a close examination of interviews in Kim Ilran and Hong Ji Yoo's (2011), and (2005) directed by Jim Butterworth, Aeron Lubarsky, and Lisa Sleeth, as well as Sebastian Junger and Tim Hetherington's (2010). By a careful review, use of dissolve, cross-cutting, and flashback will be analyzed to show how different editing techniques have an impact on subjective views of the director.