Tweet
Whisper Enhancer
(Forensics Version Only)
The Whisper Enhancer dialog box
Forensics audio situations sometimes arise with multiple parties talking on a recording in which one of them sounds quite far away from the recording microphone compared to the others. This could involve two parties where one of them is at a far distance and the other one being close. Or three parties could be talking clearly to one another, but a third party, adding to the conversation, is at some relatively large distance from the recorder microphone. Another similar situation can occur wherein a completely different conversation may be occurring at another table or location and the investigator is interested in trying to reveal some of that ancillary conversation. Another situation could involve potential gunshots or screaming (or other potentially important events) occurring at a far distance from the recorder and also occurring during breaks in the primary parties conversation. Additionally, the flip of a switch from a flight data recording may be brought to the foreground using the Whisper Enhancer. The Diamond Cut Whisper Enhancer is a special routine designed to bring out soft voices or sounds that are occurring amidst louder voices while softening the louder voices or sounds. It can reduce the loudness of the up-front voices while increasing the loudness of the far distant voices or sounds, so long as they are not occurring concurrently. Sometimes, it will be found that passing the audio through a speech filter (a preset under the Band Pass filter) or the Auto Voice filter first will provide better results with the Whisper Enhancer. It is important to note that this is not intended to be used as a de-noising filter and it has the capability of introducing distortion into the signal when its controls are not properly set for the audio material of interest. Here are the controls that you have to work with in the Diamond Cut Whisper Enhancer:
Threshold: This control sets the signal level at which the system begins to invoke its action.
Range: -60 dB to 0 dB (0 dB is the least aggressive; – 60
dB will yield the most aggressive setting)
Enhance Level Bar graph: Use this to aid in finding the “sweet-spot” for Whisper Enhancement. The higher that the green bar graph moves, the greater will be the action of the enhancer.
Ratio: This controls the magnitude of loudness between the loudest and softest sound processed. Higher values of ratio produce a greater “Whisper Enhancement” effect. Generally speaking, your best results will occur with ratio values set somewhere between 15 and 50 depending on the nature of your audio material.
Range: 5 to 50 (relative units of measurement)
Response Time: This determines the time delay before the system activates. Small values of response time produce very fast transient response to changes, but produce the greatest amount of distortion. Conversely, high values of response time produce slower transient response to changes, but less signal distortion.
Range: 1.0 mSec to 100 mSec
Output (Level): Adjust the signal level coming out of the Whisper Enhancer algorithm.
Range: -40 dB to +19.9 dB
Clip Led: This red indicator lights up when the output signal hits full scale (maximum = 0 dB). Use the Output (Level) control to reduce the signal level until this indicator extinguishes.
Bypass: Defeats the Whisper Enhancer for signal comparison purposes.
Keep Residue: The difference between the input and processed signal.
Presets: 15 Descriptive Factory Presets are provided; start with the one that best describes your audio issue and then fine-tune the various parameters for optimal whisper enhancement. After you find the best one, you can “Save” it for future reference using a name of your choosing.
WARNING: THE WHISPER ENHANCER CAN PRODUCE EXTREMELY HIGH SOUND VOLUME LEVELS WHEN NOT PROPERLY ADJUSTED. KEEP YOUR AUDIO SYSTEM VOLUME LEVEL SET TO A LOW VALUE WHEN STARTING OUT WITH THIS ROUTINE. ALWAYS BEGIN A PROJECT WITH THE OUTPUT LEVEL SET TO – 40 dB AND ADJUST TO A REASONABLE LEVEL THEREAFTER.
Note: The Whisper Enhancer works best with files that have been converted to a 44.1 kHz sample rate & having 16 bit depth (resolution).
(Forensics Version Only)
The Whisper Enhancer dialog box
Forensics audio situations sometimes arise with multiple parties talking on a recording in which one of them sounds quite far away from the recording microphone compared to the others. This could involve two parties where one of them is at a far distance and the other one being close. Or three parties could be talking clearly to one another, but a third party, adding to the conversation, is at some relatively large distance from the recorder microphone. Another similar situation can occur wherein a completely different conversation may be occurring at another table or location and the investigator is interested in trying to reveal some of that ancillary conversation. Another situation could involve potential gunshots or screaming (or other potentially important events) occurring at a far distance from the recorder and also occurring during breaks in the primary parties conversation. Additionally, the flip of a switch from a flight data recording may be brought to the foreground using the Whisper Enhancer. The Diamond Cut Whisper Enhancer is a special routine designed to bring out soft voices or sounds that are occurring amidst louder voices while softening the louder voices or sounds. It can reduce the loudness of the up-front voices while increasing the loudness of the far distant voices or sounds, so long as they are not occurring concurrently. Sometimes, it will be found that passing the audio through a speech filter (a preset under the Band Pass filter) or the Auto Voice filter first will provide better results with the Whisper Enhancer. It is important to note that this is not intended to be used as a de-noising filter and it has the capability of introducing distortion into the signal when its controls are not properly set for the audio material of interest. Here are the controls that you have to work with in the Diamond Cut Whisper Enhancer:
Threshold: This control sets the signal level at which the system begins to invoke its action.
Range: -60 dB to 0 dB (0 dB is the least aggressive; – 60
dB will yield the most aggressive setting)
Enhance Level Bar graph: Use this to aid in finding the “sweet-spot” for Whisper Enhancement. The higher that the green bar graph moves, the greater will be the action of the enhancer.
Ratio: This controls the magnitude of loudness between the loudest and softest sound processed. Higher values of ratio produce a greater “Whisper Enhancement” effect. Generally speaking, your best results will occur with ratio values set somewhere between 15 and 50 depending on the nature of your audio material.
Range: 5 to 50 (relative units of measurement)
Response Time: This determines the time delay before the system activates. Small values of response time produce very fast transient response to changes, but produce the greatest amount of distortion. Conversely, high values of response time produce slower transient response to changes, but less signal distortion.
Range: 1.0 mSec to 100 mSec
Output (Level): Adjust the signal level coming out of the Whisper Enhancer algorithm.
Range: -40 dB to +19.9 dB
Clip Led: This red indicator lights up when the output signal hits full scale (maximum = 0 dB). Use the Output (Level) control to reduce the signal level until this indicator extinguishes.
Bypass: Defeats the Whisper Enhancer for signal comparison purposes.
Keep Residue: The difference between the input and processed signal.
Presets: 15 Descriptive Factory Presets are provided; start with the one that best describes your audio issue and then fine-tune the various parameters for optimal whisper enhancement. After you find the best one, you can “Save” it for future reference using a name of your choosing.
WARNING: THE WHISPER ENHANCER CAN PRODUCE EXTREMELY HIGH SOUND VOLUME LEVELS WHEN NOT PROPERLY ADJUSTED. KEEP YOUR AUDIO SYSTEM VOLUME LEVEL SET TO A LOW VALUE WHEN STARTING OUT WITH THIS ROUTINE. ALWAYS BEGIN A PROJECT WITH THE OUTPUT LEVEL SET TO – 40 dB AND ADJUST TO A REASONABLE LEVEL THEREAFTER.
Note: The Whisper Enhancer works best with files that have been converted to a 44.1 kHz sample rate & having 16 bit depth (resolution).