John F. Kennedy assassination Dictabelt recording

Made on a common Dictaphone-brand dictation machine that recorded sound in grooves pressed into a thin vinyl-plastic belt, the recording gained prominence among Kennedy assassination conspiracy theorists following the HSCA's 1978 conclusion, based in part on this evidence, that there was a "high probability" that Lee Harvey Oswald did not act alone and that the Kennedy assassination was the result of a conspiracy.

However, after evidence from the Dictabelt recording was made available, the HSCA quickly reversed its conclusion and declared that a second gunman had fired the third of four shots heard.

In a January 5, 1980, interview with journalist Earl Golz, HSCA chief counsel G. Robert Blakey later said: "If the acoustics come out that we made a mistake somewhere, I think that would end it [the multiple-shooter theory].

Investigators compared "impulse patterns" (suspected gunshots and associated echos) on the Dictabelt to 1978 test recordings of Carcano rifles fired in Dealey Plaza from the sixth floor of the Texas School Book Depository and from a stockade fence on the grassy knoll forward and to the right of the location of the presidential limousine.

Acoustics analysts Mark Weiss and Ernest Aschkenasy of Queens College reviewed the BBN data and concluded that "with the probability of 95% or better, there was indeed a shot fired from the grassy knoll.

"[8][failed verification] Dr. James E. Barger of BBN testified to the HSCA that his statistical analysis of the impulse patterns captured on the Dallas police recordings showed that the motorcycle with the open microphone was about "120 to 138 feet" behind the presidential limousine at the time the first shot was fired.

"[10] Using an amateur film shot of the motorcade,[11] the HSCA concluded that the recording originated from the motorcycle of police officer H. B. McLain, who later testified before the committee that his microphone was usually stuck in the open position.

[27] After the FBI disputed the validity of the acoustic evidence, the Justice Department paid for a review by the National Academy of Sciences, an organization operating with a Title 36 congressional charter.

[30] Louis Stokes, a member of the United States House of Representatives who chaired the HSCA, commented that the report "raised new and serious questions about our conclusions that need to be resolved.

Dr. Barger, the HSCA's acoustics expert, when asked about this discovery and the NAS analysis, replied,Barber discovered a very weak spoken phrase on the DPD Dictabelt recording that is heard at about the time of the sound impulses we concluded were probably caused by the fourth shot.

The NAS Committee did not examine the several items of evidence that corroborated our original findings, so that we still agree with the House Select Committee on Assassinations conclusion that our findings were corroborated[32]An analysis published in the March 2001 issue of Science & Justice by Dr. Donald B. Thomas used a different radio transmission synchronization to put forth the claim that the National Academy of Sciences panel was in error.

[38] In 2003, an independent researcher named Michael O'Dell reported that both the National Academy and Dr. Thomas had used incorrect timelines because they subtracted time for repeated sections when the playback needle got stuck, but didn't realize that there were also forward skips.

"[46] In 2003, ABC News aired the results of its investigation on a program called Peter Jennings Reporting: The Kennedy Assassination: Beyond Conspiracy.

Based on computer diagrams and recreations done by Dale K. Myers, it concluded that the sound recordings on the Dictabelt could not have come from Dealey Plaza, and that Dallas Police Officer H.B.

In the March 2005 issue of Reader's Digest, it was reported that Carl Haber and Vitaliy Fadeyev were assigned with the task of digitally restoring Dictabelt 10 by Leslie Waffen from the National Archives.

Machine learning techniques like a convolutional neural network can classify and identify between different types of firearms based on the sound of a gunshot from recorded audio used in crime scene forensics.

When interviewed by author Vincent Bugliosi, the officer acknowledged that his microphone could have been stuck in the open position (he did not recall hearing any transmissions for several minutes), and could later have become unstuck after he followed the motorcade to Parkland Hospital.