AFRL Transitions CVT3 to Help Facilitate Accident Investigations

  • Published
  • By Information Directorate
  • AFRL/IF
AFRL engineers transitioned their cockpit voice recorder transcription and timing tool (known as CVT3) to both the Air Force Safety Center and the Army Combat Readiness Center. The audio processing technology is designed to facilitate accident investigations that involve audio recordings and instrument data. The software expedites the transcription and analysis of voice recordings for investigations by displaying and linking multiple audio files simultaneously so that investigators can view the files and listen to, analyze, and transcribe the audio.

Investigators can easily enter the transcription by highlighting a section of audio; the highlighted contents then display on the screen, below the audio display and in a transcription dock. Once a user has transcribed the audio files, he or she can leverage the software's simple method for generating custom-designed reports. Time formats are easily customized to support investigations. In addition, the CVT3 software provides powerful noise reduction algorithms for impulses, narrowband interference, and wideband interference. The software contains automated noise reduction capabilities and also offers users the flexibility to design custom filters.

Most of the tool's noise reduction capabilities can operate in two modes: real-time, wherein the noise reduction capability is applied during audio playback, and non-real-time, wherein the capability alters the audio clip so that the user can save the revised audio as a separate file. The CVT3 software can also analyze the time alignment between the cockpit voice recorder and the flight data recorder. If it determines the time alignment between the two is not exact, the user can stretch or shorten the audio as necessary. The software also provides speech recognition and speaker identification tools.