The Influence of Room Acoustics and Microphone Position on the Quality of Speech Recordings—A Preliminary Analysis for Building a Polish Language Reference Database

The aim of this article is to provide a detailed analysis of the impact of room acoustics and microphone configuration on the quality of Polish speech recordings. This study serves as a preparatory stage for developing a modern reference database of speech samples, intended for applications such as voice transmission quality assessment, comparative analysis of natural and synthetic speech, and detection of voice manipulation (e.g., deepfake audio). In the pilot experiment, recordings of two professional voice-over artists were carried out in three acoustic environments: an anechoic chamber, a STUDIOBOX Premium acoustic booth (RT60 < 0.15 s [RT – Reverberation Time]), and a professional recording studio with acoustic treatment (RT60 ≈ 0.35 s). The speech was captured using two Class 1 measurement microphones placed 50 cm apart, along with a Neumann U89i microphone—a standard tool in professional voice-over production. All microphones operated simultaneously to allow a direct comparison of their characteristics and their influence on the recorded material. The recordings were evaluated perceptually. A subjective listening test was conducted remotely, with 25 expert listeners—certified audio engineers with audiometrically verified hearing. Participants assessed perceived quality and preference using a standardized rating scale. All listening tests were performed using over-ear headphones to minimize the effect of ambient noise and reverberation. The results indicate that both the recording environment and the type and positioning of microphones significantly affect the consistency and quality of speech recordings. The article concludes with a set of recommendations for future projects aimed at building Polish speech databases.