User:Sibyl.ambazhachalil/sandbox

This is the shared sandbox for the article Voice Analysis.

Sibyl's Individual Proposal
I think it would be important to add some court cases that used voice recognition as key evidence. The article currently has not made any reference to how lawyers can present voice samples as part of their evidence. We can discuss the type of voice sample used as evidence as well as the verdict and how the voice sample played a role in the final decision. We can also discuss court cases in which courts decided that voice recognition is not a valid source of evidence. I think the article should include details about how voice recognition can be controversial. We can mention how people can be wrongly accused due to this. The article could also discuss the accuracy of voice recognition. We can also discuss how the NSA taped phone calls, so that they can use them for voice recognition. We could also include what kind of characteristics the voice recognition software looks for. We can have a section on just different characteristics of voice such as creaky voice, pitch, and loudness. We can discuss various aspects of voice that can be unique to an individual, and describe how it may appear on the software. The article does not seem to have any references. It does not cite any sources when they discussed various anatomical elements of voice analysis. Citations need to be added to order to ensure that the information is valid. We would need to redo the information already in the article due to the lack of references. Sibyl.ambazhachalil (talk) 23:09, 14 February 2018 (UTC)
 * Court Cases
 * Controversies
 * Voice Analysis Methods
 * Citations

''Great ideas, including the idea to check citations. Because there currently aren't any at all, it may be worth checking carefully for plagiarism too. This kind of task is not trivial -- think of it as part of the forensic linguistic process we're modelling here. A thorough job of checking for authorship is just as valid for the purposes of our course as adding a lot of new content would be. Cakers01 (talk) 16:04, 15 February 2018 (UTC)''

The lead section of the article does not mention anything from the subsections of the article. It would be helpful to add information about the subsections into the lead in order to properly introduce the article. Currently, there are no citations for the article, so I found a source from the Rutgers Library that could be used to supplement and edit the current information. The book Voice Therapy: Clinical Case Studies looks at ways to evaluate voice, and how it would be useful in a medical setting. The article heavily focuses on the medical aspect of voice analysis/evaluation, so this source could help provide citations for the information as well as ensure that the provided information is accurate. Sibyl.ambazhachalil (talk) 01:43, 3 March 2018 (UTC)
 * UPDATE

In the article section regarding Analysis Methods, most of the methods mentioned already have separate wikipedia pages so I don't think it is worthwhile to spend too much time on that. If anything, we can add more links to the other wiki pages because the current article has no citations. The book below provides some information regarding what is already in the article so we can add that to the reference list. In chapter two of this book it discusses various characteristics of voice such as loudness, pitch, resonance, etc. The book describes what the terms are and then describes how it affects your voice. In the individual proposal, I said I wanted to look at these characteristics. In the article, we can have a subsection on this. I'm sure there are other articles on these characteristics so in our article I can just give basic information on them and then link the wikipedia page that holds more detail. Sibyl.ambazhachalil (talk) 19:56, 21 March 2018 (UTC)
 * UPDATE TWO

Felicia's Individual Proposal
One thing that the article currently fails to address is the extensive range of things Voice Analysis is utilized for. It focuses almost entirely on the anatomical basis of the technology, which is presented in a detailed and easy to understand maner, but I would like to take it a step further to demonstrate how Voice Analysis is used in practice. For example, the customer service industry is now increasingly interested in voice analysis as a means for gaining business insights. Even within this industry alone, there are several ways in which voice analysis is used. This technology is in-demand in several other markets as well such as security, law enforcement, and criminology. I intend to contribute detailed information about each of its many uses within these areas. An Israel-based company called Nemesysco has received much attention in recent years over an emerging technique within the field of Voice Analysis, which they call "Layered Voice Analysis". An innovative method of lie-detection, the developers of LVA say that the technology is able to pick up on vocal-dissonance markers as a way of signaling lies. Layered Voice Analysis is still very much up and coming, but I think it would definitely be worthwhile to include information about it, as it stems directly from the original voice analysis technology. Erikssonfel (talk) 22:38, 17 February 2018 (UTC) As suggested by Professor Akers, I will opt out of tackling ALL of the specific uses of voice analysis, and instead focus on only a select few (one or two). To these, I will go into an in-depth explanation of the logistics of voice analysis's use. The uses that I decide not to discuss I will add to the talk page as potential topics to cover by future contributors. Erikssonfel (talk) 20:14, 28 February 2018 (UTC)
 * Wide Scope of Use
 * Layered Voice Analysis
 * REVISION - FEBRUARY 28TH

Mainspace Information
Characteristics of Voice

Some characteristics of voice are phonation, pitch, loudness, and rate. These characteristics can be used to evaluate a person's voice and can aid in the voice analysis process. Phonation is typically tested by looking at different types of data collected from a person such as words with long vowels, words with many phonemes, or just typical speech. A person's pitch can be evaluated by making the person produce the highest and lowest sounds they can, as well as sounds in between. A keyboard can be used to aid in this process. Loudness is valuable to look at because for certain people, loudness affects the way they produce certain sounds. Some people need to speaker louder for certain phonemes in comparison to others just so they can produce them. This can be tested by asking the person to use the same amount of loudness while singing a scale. Rate is also important because it looks at how fast or slow a person speaks. (Information is from Voice Therapy book cited above) Sibyl.ambazhachalil (talk) 17:35, 22 March 2018 (UTC)

Use in Forensics

Forensic voice analysis is currently being utilized in a broad range of domains, including criminal cases that involve murder, rape, drug dealing, bomb threats, and terrorism. Oftentimes voice might serve as the sole clue for police and forensic analysts in identifying criminals, which has led to voice analysis becoming one of the newest and most developing branches of this old science of analyzing crime scenes. Voice identification is a particularly predominant component of voice analysis, which uses a specially designed software that takes the recording in question, as well as a recording of a known person's voice and then compares the two utilizing a series of three tests. Typically, a recording of at least seven seconds is required for optimally accurate results to be achieved. The software conducts a spectrograph analysis, followed by an average pitch analysis, and lastly a statistical analysis that includes a compiled database with millions of voices. After running the voice sample through the program, a percentage from 0 to 100 is generated as far as the likelihood that the two voices are the same. The forensic analyst will then make comparisons on the accent, syntax, and breathing patterns of the recordings, an analysis that the software is not yet able to make. This voice analysis technique has gained much notoriety for its use in the Trayvon Martin case, where a recording of a call made to the police was analyzed to determine if background screams came from George Zimmerman or from Martin. Erikssonfel (talk) 15:18, 23 March 2018 (UTC)

Mainspace Information (REVISED after peer review)
Use in Forensics

Forensic voice analysis is currently being utilized in a broad range of domains, including criminal cases that involve murder, rape, drug dealing, bomb threats, and terrorism. Oftentimes voice might serve as the sole clue for police and forensic analysts in identifying criminals, which has led to voice analysis becoming one of the newest and most developing branches of this old science of analyzing crime scenes. Voice identification is a particularly predominant component of voice analysis, which uses a specially designed software that takes the recording in question, as well as a recording of a known person's voice and then compares the two utilizing a series of three tests. Typically, a recording of at least seven seconds is required for optimally accurate results to be achieved. The software conducts a spectrograph analysis, followed by an average pitch analysis, and lastly a statistical analysis that includes a compiled database with millions of voices. After running the voice sample through the program, a percentage from 0 to 100 is generated as far as the likelihood that the two voices are the same. The forensic analyst will then make comparisons on the accent, syntax, and breathing patterns of the recordings, an analysis that the software is not yet able to make. This voice analysis technique has gained much notoriety for its use in the Trayvon Martin case, where a recording of a call made to the police was analyzed to determine if background screams came from George Zimmerman or from Martin. Erikssonfel (talk) 19:35, 4 April 2018 (UTC)

Characteristics of Voice Some characteristics of voice are phonation, pitch, loudness, and rate. These characteristics can be used to evaluate a person's voice and can aid in the voice analysis process. It can help pinpoint differences in a person's voice in comparison to another person. Phonation is typically tested by looking at different types of data collected from a person such as words with long vowels, words with many phonemes, or just typical speech. A person's pitch can be evaluated by making the person produce the highest and lowest sounds they can, as well as sounds in between. A keyboard can be used to aid in this process. Keyboards are useful because the person can try to mimic the sounds made by the keyboard, especially when they near the end of the scales. Loudness is valuable to look at because for certain people, loudness affects the way they produce certain sounds. Some people need to speaker louder for certain phonemes in comparison to others just so they can produce them. This can be tested by asking the person to use the same amount of loudness while singing a scale. For some people, they will be unable to produce the sounds once they reach a certain note. Rate is also important because it looks at how fast or slow a person speaks. (Information is from Voice Therapy book cited above) Sibyl.ambazhachalil (talk) 22:08, 5 April 2018 (UTC)