Voice Alteration, Electronic

views updated Jun 08 2018

Voice Alteration, Electronic

In most cases, voice alteration technologies are employed to obscure an individual's identity. The ability of to alter the voice, however, also can be very useful in intelligence gathering and espionage. Impersonating a target individual (e.g., a worker important in the hierarchy of an organization) can provide access to information that is privy to only a select group of people. As well, if the voice alteration is sufficiently close to the original, access to files or physical locations that are barred by voice recognition software can be granted.

Crude voice alteration can be achieved by physical training. Actors and singers, for example, can train their voices so that the speech or song "projects" to all areas of the theatre. Also, accents can be learned and mimicked with reasonable accuracy.

In this natural process the vocal cords function as the source of the sounds and the vocal tract functions as the filter that can alter the frequency and cadence of the speech. The results is the rising and falling tones and intensity of spoken words.

However, the use of electronic technology can achieve accurate vocal alterations that are not otherwise possible. For example, vocal cords can be trained to be able to adopt different pitches—that is, to be capable of vibrating at different frequencies, so as to produce sounds that have different tones. However, electronic alterations of pitch can widen the vocal deceptions that are possible. For example, a man's voice can be altered to sound absolutely convincingly like a woman's.

The alteration of pitch can also be deliberately done electronically by detecting the frequency pattern of the speaker, and of the particular phrase being spoken. On a screen, the pattern appears as a series of waves and troughs. The arrangement of the waves and troughs is characteristic to the word being spoken. For example, the word "cat" will produce a different pattern than the word "invisible." By applying an electronic filter (or "window"—actually one or more mathematical equations, or algorithms) to the frequency pattern, waves and troughs can be selectively eliminated or shifted up and down to produce a different frequency. An experienced technician or sophisticated software program can alter a word so as to change the sound of the word (i.e., a higher or lower tone) without distorting the sound of the word. Thus, the altered speech is still recognizable and interpretable, but can sound like it is being spoken by another person.

Electronic voice alteration can be subtle or extreme. The latter is associated with the almost incomprehensible voices of anonymous witnesses. This type of voice alteration is actually a voice disguise. The intention is not to mimic a voice, but to scramble the voice patterns to make the identify of the speaker impossible to identify.

There are several different electronic means of voice alteration. One type is known as speech inversion. Here, the frequency signal is in effect turned inside out around a designated frequency. Put another way, the parts of the speech that are "high" are made to sound "low", and visa versa.

A voice can also be electronically jumbled, so that it sounds like gibberish. But codes assigned to sections of the speech allows the listener (who has the electronic codes) to put the words back in their proper order.

Another means of electronic voice alteration is known as speech encryption. Here, speech is digitized and the digital signal manipulated to make the text of the speech unrecognizable to the listener's ear. But, the speech can be decoded, or decrypted, at the receiving end to yield the original recognizable speech.

Hardware and software voice encryption systems are available. Machines connected to a telephone can alter a person's speech during the telephone conversation. Anyone eavesdropping on the conversation would be incapable of understanding what was being said. However, a legitimate listener, having a machine on his or her phone, would be capable of decrypting the conversation.

The United States government and military uses a telephone conversation scrambling software program and hardware called STU III (Secure Telephone Unit, Generation III).

Scrambling digital electronic information in relation to time can also accomplish voice alteration. An example includes the delay of information. While an effective means of altering a voice, the method can produce an echo, and so is unpleasantly distracting to listen to.

█ FURTHER READING:

BOOKS:

Hollien, Harry Francis. The Acoustics of Crime: The New Science of Forensic Acoustics. New York: Plenum Press, 1990.

——. Forensic Voice Identification. New York: Academic Press, 2001.

Voice Alteration, Electronic

views updated May 09 2018

Voice Alteration, Electronic

In most cases, voice alteration technologies are employed to obscure an individual's identity. This can complicate the forensic identification of the individual. Forensic specialists who examine spoken or written materials in relation to legal matters and crimes are known as forensic stylists or forensic linguists. Forensic linguists frequently deal with evidence containing altered voices. Crude voice alteration can be achieved by physical training. Actors and singers, for example, can train their voices so that the speech or song "projects" to all areas of the theater. Also, accents can be learned and mimicked with reasonable accuracy.

However, the use of electronic technology can achieve accurate vocal alterations that are not otherwise possible. For example, vocal cords can be trained to be able to adopt different pitches—that is, to be capable of vibrating at different frequencies, to produce sounds that have different tones. However, electronic alterations of pitch can widen the vocal deceptions that are possible. For example, a man's voice can be altered to sound convincingly like a woman's.

The alteration of pitch can also be deliberately done electronically by detecting the frequency pattern of the speaker, and of the particular phrase being spoken. On a screen, the pattern appears as a series of waves and troughs. The arrangement of the waves and troughs is characteristic to the word being spoken. For example, the word "cat" will produce a different pattern than the word "invisible." By applying an electronic filter (or "window"; actually one or more mathematical equations, or algorithms) to the frequency pattern, waves and troughs can be selectively eliminated or shifted up and down to produce a different frequency. An experienced technician or sophisticated software program can alter a word so as to change the sound of the word (i.e., a higher or lower tone) without distorting the sound of the word. Thus, the altered speech is still recognizable and interpretable, but can sound like it is being spoken by another person.

A voice can also be electronically jumbled, so that it sounds like gibberish. But codes assigned to sections of the speech allow the listener (who has the electronic codes) to put the words back in their proper order.

Another means of electronic voice alteration is known as speech encryption. Here, speech is digitized and the digital signal manipulated to make the text of the speech unrecognizable to the listeners ear. But the speech can be decoded, or decrypted, at the receiving end to yield the original recognizable speech.

The United States government and military uses a telephone conversation scrambling software program and hardware called Secure Telephone Unit, Generation III (STU III).

see also Linguistics, forensic stylistics; Telephone recording system; Telephone tap detector.

World of Forensic Science