We met again as planned to record the sounds of the monument and the banknote. First, at the Hirobumi Ito monument at UCL, we attempted to record using a contact microphone and a Sennheiser condenser microphone. Since the monument was made of stone, it had very little resonance, making it quite difficult to capture sound. We wanted to record a wide range of sounds, from subtle noises like gentle taps and rubbing to more intense impacts, but very faint sounds were hard to capture even with the contact mic. To address this, we recorded sounds by rubbing the contact mic itself against the monument and striking it with more force. For the condenser microphone, we covered it and rubbed it directly against the stone to capture friction-based sounds. Our main focus was not just on capturing general “stone” sounds but rather on extracting unique sonic characteristics specific to this particular monument. Therefore, we did not hesitate to use sounds that might not typically be associated with stone.

For the banknote, we also used both the contact mic and the condenser mic, recording in an indoor space that was as quiet as possible. Unlike the stone, the banknote easily produced sound, and the recording process itself was relatively straightforward. However, the contact mic was not the most suitable choice for this material. As the banknote moved, micro-gaps continuously formed between it and the contact mic, generating low-frequency noise. Fortunately, the condenser microphone was well-suited for capturing clean audio, allowing us to obtain high-quality recordings. That said, the contact mic still produced an interesting and unique sound, so we plan to experiment with it during post-production.

With this, we have completed the recording phase. After processing and refining the recorded sounds, we will decide whether additional recordings are necessary. We have decided to use a vocoding technique to overlay the recorded sounds onto human speech. In addition, we plan to experiment with AI-generated voices, historical recordings of Hirobumi Ito from media sources, and even manually arranging sounds to mimic human intonation without vocoding. By testing these different approaches, we aim to determine which method best suits our project.