Abstract Background: There is always an interaction between man and his environment that can be the cause of physical, physiological and psychological stress on people and also cause discomfort, annoyance, and have direct and indirect effects on their performance and productivity, health and safety. People in their workplace are exposed to many factors related to work activities and environmental factors, which can affect their health; So workplace disruptions such as noisy environments can cause safety problems, reduce employee morale and undesirable effects on performance, working memory, and attention; For these reasons, ergonomic designers must be capable of designing the physical environment because the design of the environment not only keep peoplechr('39')s performance high, but also provide a safe and healthy environment for improving employeeschr('39') health. Noise can also negatively affect communication between people. Banks are open plan offices and the financial services provided by bank staff that requires employeechr('39')s mental activity. Different sources of noise production can affect employees and negatively affect their comfortable, functional and communication aspects. Noise also effects on mental activities can cause employees to become tired early and increase their computational errors. In a bank environment where conversations between bank tellers and customers are necessary to conduct banking tasks, interfering with ambient noise, causing annoyance, human error, and ultimately failed to provide banking services. Because of high workload and the bankschr('39') competition to attract customers in order to provide better services, quiet workplace environments are essential for staff comfort and optimum performance; so right designing workplace can have a high impact on peoplechr('39')s comfort and productivity. Speech Interference Level is a simple way to predict or evaluate speech intelligibility where the conversation occurs directly in a noisy environment. At open plan offices, employee and customer communicate face-to-face and information exchange is done in that way. In recent years, many banks in our country have established their own workstations as sited customers and face-to-face with bank tellers, but so far no study has been conducted on the changed employeeschr('39') workstations and its effect on speech interference. A question now arises, is workstation redesign increase the Speech Interference Level and improve Intelligibility ratings for speech communications? Therefore, the purpose of this study was redesigned workstations and make environmental changes and its effect on SIL among bank tellers. Material and method: This study is a descriptive-analytic and Interventional study, this was carried out for one year in three branches of one of the governmental banks of Hamadan province (N= 12). Twelve workstations were redesigned and installed after measuring the sound and calculating Speech Interference Level, and calculating the distance between customer and employee. Bank tellers are those who have a direct relationship with customers and perform banking tasks. They had to sit long during the day and they spent all their time working on computers and they were getting things done and receiving customer financial requests. They used the workstations were constructed in the present study. The tools for data collection in this study were: 1) demographic and occupational questionnaire, 2) sound measuring and frequency analyzer CEL-450، 3) A questionnaire with a question “What noise sources in your workplace do you annoy?” to determine annoying sound sources and 4) ISO 9921-2003 standard to determine speech interference level of the noise at the listenerchr('39')s ear (LSIL), Speech level, Speech Interference Level (SIL) and Intelligibility ratings for speech communications. SIL is a simple way to predict or evaluate speech intelligibility, Speech intelligibility also increases with increasing SIL. speech intelligibility is divided into 5 rating (bad, poor, fair, good and excellent) based on SIL, also in sensitive situations where short messages are sent include important numbers, ability to understand at least "good" is recommended by increasing speech effort; the communication will be “fair” if the difference between LSIL and the speech level (LS, A, L) that both determined at the listener’s position is greater than 10. Measuring the sound within 1 meter of the speakerchr('39')s mouth, according to ISO 9921 (2003) for normal sound (60 dB) was considered. It was measured for 15 minutes at 10 a.m. to 13 p.m. every shift that they had the highest number of customers, and measuring was done twice for each workstation (total 24 times). To calculate LSIL, the sound pressure level was measured in four octave bands with the central frequencies 500 Hz, 1000 Hz, 2000 Hz and 4000 Hz at the listener position and during the normal communication period in time-weighting “slow”. the sound pressure level was measured for 15 minutes and the average of these pressure levels was calculated. After collecting the preliminary data, bank teller workstations were redesigned and built. Some actions were taken to reduce speech interference and improve communication between bank tellers and customers; For example, insert a glass between customer and bank teller and create 12 vents (diameter =2 cm) at a customerchr('39')s mouth height to improve communication, changing the workstation so that the operator and customer interact face-to-face and insert chair for customers to sit in, reduce customer gathering in front of the workstations, reducing the distance between the customers and the bank tellers by creating an arc at the table, Enhancing employee privacy by placing the glass between employees in beside workstations, increasing the glass height between the bank teller and the customer to prevent customer standing communicate with the bank teller. The mean distance between bank teller and customer prior to redesigning the workstations was 123 cm which was reduced to 103 cm in redesign; the mean distance between two employees was 95 cm, which increased to 115 cm after the redesign. Environmental interventions also include the installation of perforated acoustic tiles made of plaster as a false ceiling, so that, by installing them the distance between the main and false ceilings was 25 cm and there was an empty space between them which absorbs more sound. The walls of the branches were also covered with wall plugs that made of PVC and MDF, and the floor was improved. Data were collected before and after the interventions and they were compared. Data analysis was performed using SPSS software, version 16. Data were analyzed using descriptive statistics (mean and standard deviation) and analytical statistics. The Paired t-test was used to compare the mean results of the LSIL and SIL. Results: In the present study, all the bank tellers were male. Everyone was married, their mean age was 38.25 (5.62) and work experience was 15.25 (6.87) years. Each bank teller were doing 30-50 customerschr('39') finances task in one shift. All participants stated that their rest time was less than 30 minutes in a shift and they are working 7-8 hours in a sitting position; none of the participants had a history of hearing loss or other hearing problems. Results from annoying sources of noise showed that the most annoying sources of noise in all three branches were the noise of customers (42.85%) and the sound of money counting machine (33.3%); Printer noise was the lowest (4.76%). Before the interventions, mean speech interference level (LSIL) was 59.50 dB (4.07) that was then reduced to 54.98 (5.52) dB and this decrease was statistically significant (p