The influence of recording conditions on the results of acoustic analysis

The second test was to examine the technical parameters of recording devices. Due to the anticipated use of the system to carry out mass tests, both specialist equipment, as well as generally available and cheap equipment were tested. Specialist equipment included:

• specialized M-Audio MobilePRE audio interface with built-in microphone preamplifier and power supply for a condenser microphone,
• Behringer Mic200 preamplifier,

and the tested popular class equipment is:

• built-in notebook sound card (Realtek chipset),
• treated as "better" - Creative SoundBlaster External Live!

The measuring microphone in the case of sound cards without the possibility of powering a condenser microphone was a higher-class dynamic vocal microphone (XM8500) and a cheap computer electret microphone, and in the case of using an audio interface - a Behringer B1 condenser microphone.

The Yamaha HS50M studio monitor with a relatively "flat" frequency response was used as the sound source.

Methodology

In order to initially define expectations regarding the results, the research began with tests of the sound cards themselves in the so-called "Loops", i.e. without the use of microphones and loudspeakers, using the line inputs of the cards. Due to unrealistically good results, they were not taken into account in the next stages of work and will not be presented. It seems possible that manufacturers of sound cards disconnect amplifiers at too low a signal level, achieving with this virtual method better dynamics of the equipment during this type of research. Therefore, an additional test of the entire audio path was carried out, in which the replaced element was the microphone, amplifier and card. Four sets were created for the test:

• a computer microphone connected to the built-in sound card,
• a vocal microphone connected to the SB Live!,
• a vocal microphone connected via a preamplifier to the line input of the built-in sound card,
• studio condenser microphone connected to a professional audio interface.

They were composed in this way because we considered it advisable to check whether a better combination is a good microphone and the so-called "Better" sound card, or a good microphone with a preamplifier and an average sound card. The recordings were made in a quiet anechoic chamber, but without the elimination of electromagnetic interference. The microphone was placed at a relatively large distance (approx. 25 cm) from the loudspeakers to remove the effects of "boosting" the bottom of the dynamic microphone's frequency response.

Bandwidth

The bandwidth was checked first. Results are shown on Fig.4. What points to your attention is definitely the widest and most balanced frequency response of a set with a condenser microphone and a professional audio interface.

In the case of a dynamic microphone, it becomes noticeable that the transmission characteristics are limited from the bottom, especially in cooperation with the built-in sound card. A cheap computer microphone has significantly better characteristics in the lower frequency range than it, and even when attached to the built-in card, it performs slightly better than a dynamic vocal microphone connected to a dedicated preamplifier. In the high-frequency range, a studio condenser microphone is definitely the best. A dynamic microphone connected to a preamplifier also provides a frequency response of over 10,000 Hz with a flat response. Unfortunately, the other combinations, and especially the set consisting of a computer microphone connected to the built-in card, show a significant deterioration of the characteristics. A computer microphone connected to the microphone input of the built-in card starts to decrease its sensitivity from around 5000 Hz, which may have a strong impact on later analysis results. This phenomenon was not observed when an additional external preamplifier connected to the line input was used with the same board. Subsequent research and analysis of the loop test results for this card showed that the main cause was the internal microphone preamplifier on the board.

Signal to noise ratio

The results of the signal-to-noise ratio tests are shown in Fig. 5. The remarkably high level of spacing achieved for the audio interface indicates a remark. Probably the low noise level of the studio microphone disconnects the input amplifiers in this case. This kind of "distortion" unfortunately disqualifies this type of test. It is required to bypass the hardware "improvement", which should be easily achieved in the signal dynamics test.

Dynamics

The results of the dynamics tests are presented in Fig. 6. As can be seen in the graphs, due to the presence of the reference signal with a frequency of 1000 Hz (the characteristic "peak" in the middle of the graph), the input amplifiers were not disconnected in any case.

In low frequencies, a cheap computer microphone does not work very well. Due to its high sensitivity in low frequencies and poorly shielded, unbalanced connection cable, the network noise is unfortunately quite strong. In the lowest frequency range, also in the case of the potentially best studio microphone setup (and only in this case), there are some disturbing components. Their presence is partly due to the extremely low cut-off frequency of this set (according to the manufacturer - 20 Hz), and to some extent from the presence of delicate low-frequency interference from the air conditioning system operating in the neighboring room, which could not be eliminated. The hum in this case is relatively low and related rather to the parameters of the preamplifier built into the interface. This indicates special requirements for the use of this equipment. The vocal microphone connected to the preamplifier was chosen as the best set in low frequencies. Since in this case the sound card is an integrated card, it additionally indicates that the source of interference in the case of a computer microphone + built-in card is the microphone.

In the high frequency range, a studio microphone performs best, despite the broadest frequency response and highest sensitivity. The worst set is the SB Live! and a vocal microphone, if not for the noted decrease in the sensitivity of the computer microphone connected to the microphone input of the built-in card. This suggests that the card manufacturer has deliberately lowered the high-frequency sensitivity to eliminate some of the noise. In the case of SB Live! this modification was not made to such an extent, which ultimately resulted in a high level of noise.

Conclusions

The best setup is without a doubt a studio microphone with a professional audio interface. It is important, however, that it is close to a set consisting of a good vocal dynamic microphone with a dedicated preamplifier connected to the line input of the sound card. This set is also safer to use, and as long as the registration of the lowest frequencies is not required (e.g. "prosthetic" voices, in patients with inserted voice valves after laryngeal removal procedures), it will fulfill its functions perfectly. However, the set with a slightly "better" sound card did not work.