Multi-feature speech/music discrimination system
2010-03-29 00:00:00
the time window, the variance will be relatively high. In contrast, music is likely to be constant within the time window, and therefore will have a lower variance value. The differences in the variance values can therefore be employed to distinguish speech sounds from music. By combining data from some of the base features with data from other features, such as the variance features, significant increases in the discrimination accuracy are obtained.

In another aspect of the invention, a "nearest-neighbor" type of classifier is used to distinguish speech data samples from music data samples. Unlike the Gaussian classifier, the nearest-neighbor classifier estimates local probability densities within every area of the feature space. As a result, arbitrarily complex decision boundaries can be generated. In different embodiments of the invention, different types of nearest-neighbor classifiers are employed. In the simplest approach, the nearest data point in the feature space to a sample data point is identified, and the sample is labeled as being of the same class as the identified nearest neighbor. In a second embodiment, a number of data points within the feature space that are nearest to the sample data point are determined, and the new sample point is classified by a voting technique among the nearest points in the feature space. In a preferred embodiment of the invention, the number of nearest data points in the feature space that are employed for such a decision is small, but greater than unity.

In a third embodiment, a K-d tree spatial partitioning technique is employed. In this embodiment, a K-d tree is constructed by recursively partitioning the feature space, beginning with the dimension along which features vary the most. With this approach, the decision boundary between classes can become arbitrarily complex, in dependence upon the size of the set of features that are used to provide input data. Once the feature space is divided into sufficiently small regions, a voting technique is employed among the data points within the region, to assign it to a particular class. Thereafter, when a new sample data point is generated, it is labeled according to the region within which it falls in the feature space.

The foregoing principles of the invention, as well as the advantages offered thereby, are explained in greater detail hereinafter with reference to various examples illustrated in the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS:

FIG. 1 is a general block diagram of a speech/music discriminator embodying the present invention;

FIG. 2 is an illustration of an audio signal that has been divided into frames;

FIGS. 3a and 3b are histograms of the spectral centroid for speech and music signals, respectively;

FIGS. 4a and 4b are histograms of the spectral flux for speech and music signals, respectively;

FIGS. 5a and 5b are histograms of the zero-crossing rate for speech and music signals, respectively;

FIGS. 6a and 6b are histograms of the spectral roll-off for speech and music signals, respectively;

FIGS. 7a and 7b are histograms of the cepstral resynthesis residual magnitude for speech and music signals, respectively;

FIG. 7c is a graph showing the power spectra for voiced speech and a smoothed version of the speech signal;

FIGS. 8a and 8b are graphs depicting variances between speech and music signals, in general;

FIGS. 9a and 9b are histograms of the variation in spectral flux for speech and music signals, respectively;

FIGS. 10a and 10b are histograms of the proportion of low energy frames for speech and music signals, respectively;

FIG. 11 is a block diagram of a speech modulation detector;

FIGS. 12a and 12b are histograms of the 4 Hz modulation energy for speech and music signals, respectively;

FIG. 13 is a block diagram of a circuit for determining the pulse metric of signals, along with corresponding signal graphs for two bands at each stage of the circuit;

FIGS. 14a and 14b are histograms of the pulse metric for speech and music signals, respectively;

FIG. 15 is a graph illustrating the probability distributions of two measured features;

FIG. 16 is a more detailed block diagram of a discriminator; and

FIG. 17 is a graph illustrating an example of speech/music decisions for a sequence of frames.

DETAILED DESCRIPTION

In the following discussion of various embodiments of the invention, it is described in the context of a speech/music discriminator. In other words, all input sounds are considered to fall within one of the two classes of speech or music. In practice, of ...

Musical scale indicator
2010-03-26 00:00:00
from which all music derives, is based upon the natural tones, A, B, C, D, E, F, and G (the white piano keys), as well as upon the accidentals F鈾? G鈾? A鈾疌鈾?and D鈾?(the black piano keys).

In any melody, there is one tone which seems to dominate and be more final than any other tone. If a musical melody is played without finishing on this tone, the melody appears to the ear as somehow incomplete. This central tone is called the "tonic", or "key ". Each tonic has a set of tones which are related to it in varying degrees. When a musical score begins on a certain tone, it can be expected that certain selected tones will follow. These groups of tones, which relate to the concept of "tonality", constitute the musical "scales".

Over the years a number of musical scales have been developed, as follows.

By the seventeenth century, the following scales (or modes) were in use:

A B C D E F G A ... known as Aeolian;

B C D E F G A B ... known as Locrian;

C D E F G A B C ... known as Ionian;

D E F G A B C D ... known as Dorian;

E F G A B C D E ... known as Phrygian;

F G A B C D E F ... known as Lydian; and

G A B C D E F G ... known as Mixolydian.

The Major scale, like Ionian Mode, is based upon a succession of eight tones modeled on the tone intervals, or steps, when the succession of tones begins on C. These intervals are: C-D, D-E, E-F, F-G, G-A, A-B, and B-C; constituting steps which are: whole, whole, half, whole, whole, whole, and half. This scale is known as the "C Major Scale". Major scale beginning on other tones may be constructed, always with the steps between the third and fourth tones and the seventh and eighth tones being half steps. This is accomplished by selectively utilizing the accidentals A鈾? B鈾? C鈾? D鈾? E鈾? F鈾痑nd G 鈾? and Ab Bb, Db, Eb, Fb and Gb instead of the naturals A, B, C, D, E, F, and G, as necessary to achieve the intervals, or steps, of the C Major For instance, the Major Scale beginning on G is constructed as scale. For instance, the Major Scale beginning on G is constructed as follows: G A B C D E F 鈾疓.

The Minor Scale, like the Aeolian Mode, is based upon a succession of eight tones modeled on the tone intervals, or steps, when the succession of tones begins on A. These intervals are: A-B, B-C, C-D, D-E, E-F, F-G, and G-A; constituting steps which are: whole, half, whole, whole, half, whole, and whole. As in the Major Scale, the Minor Scale can be constructed so as to begin on any tone with the intervals between tones being those of A minor, by using the appropriate accidentals of the tones where required.

The foregoing Minor Scale description is known as the "Natural Minor Scale". There are two main variations of the Minor Scale. The "Harmonic Minor Scale" is an adaptation of the Minor Scale for harmonic purposes in certain melodies. The Harmonic Minor Scale raises the seventh step so that there is a half-step difference between the seventh and eighth steps of the octave. The intervals are: A-B, B-C, C-D, D-E, E-F, F-G鈾? and G鈾?A; constituting steps which are: whole, half, whole, whole, half, one and one half, and half. The "Melodic Minor Scale" additionally raises the sixth step when the melody is ascending, but the sixth and seventh degrees are restored to the natural when the melody is descending. The intervals when ascending are: A-B, B-C, D-E, E-F鈾? F-G鈾? and G鈾?A; constituting steps which are whole, half, whole, whole, whole and half.

A "scale" a sequential series of tones which is established under the principle of tonality. In contradistinction to this is the concept of the "chord", which is the simultaneous playing of more than one tone.

There are four basic families of musical instruments: stringed, brass, woodwinds and percussion. In each family, individual instruments have unique fingerboard positions which are required in order for the...

Musical apparatus using multiple light beams to control musical tone signals
2010-03-25 00:00:00
in conjunction with the accompanying drawings.

In one embodiment of the present invention, a musical apparatus which detects light rays and uses results of this detection to control musical tones may comprise a plurality of light emitters, a single light detector, and a controller for controlling parameters of musical tone. The light emitter may be a light emitting element such as an infrared light-emitting diode (infrared LED), and a plurality of light emitters may use, for example, two infrared LEDs. Likewise, a light detector may use, for example, a light receiving element such as an infrared sensor. The plurality of light emitters and the single detector are mounted on the main housing of the apparatus. The single detector detects the light rays, which were radiated from the plurality of light emitters and reflected off of a material object in space, independently for every light emitter, and outputs the results detected corresponding to each of the plurality of light emitters, respectively. In response to the detected results, the controller controls or changes parameters of a musical tone.

In this embodiment, the plurality of light emitters are, for example, positioned at a prescribed distance (see FIG. 2), or they are positioned such that the direction of radiation of the light emitted from one light emitter is different than that of another light emitter (see FIG. 16), so that when the position of a material object is changed, the light reflected off the material object also changes. The plurality of light emitters emit light in a time-sharing manner, and the single detector outputs the detection result corresponding to the light emitter which emitted light rays synchronously with the timing of the light emission. The musical apparatus may output a detection result corresponding to each one of the plurality of light emitters respectively.

An alternative embodiment of the present invention includes a plurality of light detectors where at least one detector outputs detection results with respect to a plurality of light emitters. For instance, an embodiment of the present invention may include an apparatus containing three light emitters and two detectors where one of the two detectors outputs detection results with respect to two or three emitters. Accordingly, there is no need for a 1:1 correspondence of light detectors to light emitters, which reduces costs.

In another embodiment of the present invention, the musical apparatus which detects light rays and uses the results of this detection to control musical tones may comprises one light emitter, a plurality of detectors, and a controller for controlling parameters of a musical tone. The single light emitter and the plurality of detectors are mounted on the main housing of the apparatus. The plurality of detectors detects the light rays respectively, which were radiated from the single light emitter and reflected off of a material object in space, and outputs the results detected. The controller changes parameters for a musical tone based on the detection results.

In this embodiment, a plurality of detectors are, for example, positioned at a predetermined distance, or they are positioned so as to provide differing directivity thereof in their detection regions from one another, so that when a position of the material object in space is changed, a condition in detecting the light reflected by the material object changes.

In the musical apparatus containing a plurality of light emitters, at least one light emitter is noticed in the sense that the light radiated from the light emitter is detected by a plurality of detectors. For instance, in an apparatus containing three light emitters and two detectors, one of the three light emitters may be noticed in the sense that the light rays from the light emitter is detected by two detectors. Accordingly, there is no need of a 1:1 correspondence of light emitters to light detectors, which reduces costs.

In these embodiments, the musical apparatus may further comprise a selector capable of selecting a desired parameter in a plurality of parameters, and the controller controlling changing modes of parameters selected by the selector in response to the detection results of the detectors.

In addition, the musical apparatus may still further comprise a performance mode for controlling or changing parameters of musical tones based on the detection results of a light detector, a setting mode for setting this performance mode, and a controller which, in the setting mode, sets values based on the detection results, and in the performance mode, changes parameters of musical tones based on the values set during the setting mode.

FIG. 1 is a block diagram showing an electronic musical apparatus embodying the musical apparatus of the present invention where the electronic musical apparatus is constituted such that its entire operation is controlled by the use of a central processing unit (CPU) 10, and more specifically, a bus (BUS) 12 connected the CPU 10; a read-only memory (ROM) 14 storing a program and the like executed by the CPU 10; a random access memory (ROM) 16 having an area for a control table which will be described hereinafter, an area for a buffer, similar areas for executing the program by means of the CPU 10, and a working area; a sequencer 18 in which data of musical performance for a plurality of musical pieces (the expression "data of musical performance for musical pieces" will be hereinafter referred to as "musical piece performance data") and data for musical performance expressing a phrase having a shorter performance period of time than that of musical piece performance data (the expression "data for musical performance expressing a phrase" will be hereinafter referred to as "phrase performance data", and further "phrase performance data which have been stored in a built-in ROM will be referred to as "first phrase performance data", "second performance data", and "third performance data", respectively) have been stored in a built-in ROM and which reads the musical piece performance data and phrase performance data to output the same in accordance with the processing which will be described below; a sound source 20 in which setting conditions f...

Thumbrest ring adapter for musical instrument
2010-03-24 00:00:00
in those instances when the thumbrest of the instrument does not include a permanent attachment ring. The ring adapter assembly allows a support device to be attached to the musical instrument to relieve the weight on the thumb and hand of a musician playing the instrument. One embodiment of the ring adapter assembly is used with fixed position thumbrests and is adapted to be seated on a top surface of the thumbrest so that an attachment ring within the assembly faces away from the musical instrument. A top portion of the assembly may be moved up and down by an adjustment screw and includes a spring wire which extends around the thumbrest to connect the assembly to the thumbrest. Upward adjustment of the top portion increases the tension on the wire and the downforce applied to the assembly to maintain the assembly seated on the thumbrest. Another embodiment of the ring adapter assembly fixes an attachment ring to an adjustable thumbrest. The position of the ring adapter assembly on the thumbrest is adjustable to compensate for adjustments of the thumbrest in relation to the musical instrument. An attachment component for a monopod strut device is also disclosed. The attachment component is releasably attachable to the ring adapter assembly to support the musical instrument.Claims

The invention claimed is:

1. A ring adapter assembly for a musical instrument having a thumbrest which includes a horizontal projection extending outward from the instrument toward a musician and under which a thumb of the musician is conventionally placed, said assembly comprising:

a body;

attachment apparatus for connecting the body to the thumbrest without interfering with placement and position of the musician's thumb under the horizontal projection; and

an eye fixed to the body at a position above the horizontal projection of the thumbrest.

2. A ring adapter assembly as defined in claim 1, wherein the position of the horizontal projection is fixed in relation to the musical instrument, and wherein the body further includes:

a base having a bottom surface seated upon a top surface of the horizontal projection.

3. A ring adapter assembly as defined in claim 2, wherein the body further comprises:

a resilient, compressible surface bonded to the bottom surface of the base and adapted to be seated on the horizontal projection.

4. A ring adapter assembly as defined in claim 1, wherein a vertical post is fixed to a top surface of the horizontal projection of the thumbrest and a receptacle is fixed to the musical instrument to receive the vertical post and adjustably fix the position of the horizontal projection in relation to the musical instrument, and wherein:

the body further comprises an elongated horizontal surface defining a vertical opening at one end to receive the vertical post of the thumbrest and allow the horizontal surface to move along the vertical post; and

the attachment apparatus includes movement limiting means to fix the location of the horizontal surface relative to the vertical post.

5. A ring adapter assembly as defined in claim 4, wherein:

the horizontal surface further defines a threaded horizontal opening; and

the movement limiting means comprises a set screw extending through the threaded horizontal opening to contact the vertical post.

6. A ring adapter assembly as defined in claim 4, wherein:

the body further comprises a vertical surface attached to an end of the horizontal surface opposite the vertical opening; and

the eye is fixed to the vertical surface.

7. A ring adapter assembly as defined in claim 6, wherein the vertical surface and the eye extend below the horizontal surface.

8. A ring adapter assembly as defined in claim 6, wherein the vertical surface and the eye extend above the horizontal surface.

9. A ring adapter assembly as defined in claim 6, wherein:

the horizontal surface is fixed to the vertical post at a location above the receptacle; and

the vertical surface and the eye extend below the horizontal surface.

10. A ring adapter assembly as defined in claim 6, wherein:

the horizontal surface is fixed to the vertical post at a location below the receptacle; and

the vertical surface and the eye extend above the horizontal surface.

11. A ring adapter assembly as defined in claim 1, in combination with: