US5666350A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Apparatus and method for coding excitation parameters in a very low bit rate voice messaging system
Assignee: Motorola Solutions Inc
•
Filing: 1996-02-20
•
Granted: 1997-09-09
•
Expires: 2016-02-20
Abstract
Abstract An apparatus codes excitation parameters for very low bit rate voice messaging using a method that processes a voice message to generating speech parameters. The speech parameters are separated (316) to produce a first group of energy parameters and a second group of pitch and voicing parameters. Subsequently, the first group of energy parameters are encoded and compressed using a non-uniform root-mean-square scalar process (318) to create a first plurality of encoded data. Additionally, the second group of pitch and voicing parameters are compressed, encoded, and combined into a single parameter using a three slope vector encoding process (320) that creates a second plurality of encoded data. Finally, the first and second plurality of encoded data are multiplexed (322) to create a multiplexed signal for transmission, the multiplexed signal representing the voice message.
Inventors
- 👤 Jian-Cheng Huang
- 👤 Xiaojun Li
- 👤 Floyd Simpson
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-01 14:46
CPC Classifications
G; G10; G10L; G10L19/00; G10L19/02; G10L19/0212; H; H04; H04J; H04J3/00; H04J3/17
Claims Analysis
Total Claims: 11
Claim 1
Independent Claim
114 words
A method for coding speech parameters generated from a voice message, comprising the steps of: separating the speech parameters to produce a first group of energy parameters and a second group of pitch and voicing parameters; compressing and encoding the first group of energy parameters using a non-uniform root-mean-square scalar process to create a first plurality of encoded data; compressing, encoding, and combining the second group of pitch and voicing parameters into a single parameter using a three slope vector encoding process that creates a second plurality of encoded data; and multiplexing the first and second plurality of encoded data to create a multiplexed signal for transmission, the multiplexed signal representing the voice message.
Claim 2
Independent Claim
114 words
A method for coding speech parameters generated from a voice message, comprising the steps of: separating the speech parameters to produce a first group of energy parameters and a second group of pitch and voicing parameters; compressing and encoding the first group of energy parameters using a non-uniform root-mean-square scalar process to create a first plurality of encoded data; compressing, encoding, and combining the second group of pitch and voicing parameters into a single parameter using a three slope vector encoding process that creates a second plurality of encoded data; and multiplexing the first and second plurality of encoded data to create a multiplexed signal for transmission, the multiplexed signal representing the voice message.
Claim 3
Dependent Claim
25 words
The method according to claim 1 further comprising the step of: transmitting the multiplexed signal via a transmitter and a transmission channel to a receiver.
Claim 4
Dependent Claim
129 words
The method according to claim 2 further comprising the steps of: receiving the multiplexed signal at the receiver which couples the multiplexed signal to a de-multiplexer; demultiplexing the multiplexed signal creating a reproduction of the first and second plurality of encoded data; decompressing and decoding the first plurality of encoded data using an inverse non-uniform root-mean-square scalar process to recover the first group of speech parameters; decompressing and decoding the second plurality of encoded data using an inverse three slope vector encoding process to recover the second group of pitch and voicing parameters from the single parameter; passing the first group of speech parameters and the second group of pitch and voicing parameters to a synthesizer; and synthesizing a reproduction of said speech parameters to replicate said voice message.
Claim 5
Dependent Claim
59 words
The method according to claim 1 wherein the non-uniform root-mean-square scalar process comprises the steps of: performing a discrete cosine transform on the first group of energy parameters resulting in a transformed first group of energy parameters; and quantizing the transformed first group of energy parameters using a non-uniform scalar quantizer to create the first plurality of encoded data.
Claim 6
Dependent Claim
170 words
The method according to claim 1 wherein the three slope vector encoding process comprises the steps of: assigning an unvoiced frame to a first pitch value and a voiced frame to a second pitch value; comparing a pitch value of a subsequent frame to a pitch value of a prior frame to determine a pitch trend, the pitch trend comprising a constant pitch for a subsequent frame having a like pitch value, an increasing pitch for a subsequent frame having a greater pitch value, and a decreasing pitch for a subsequent frame having a lesser pitch value; accumulating a number of like pitch trend frames until the pitch trend changes; mapping the pitch trend and the number of like pitch trend frames to a segment variable encoding the segment variable using a Huffman encoding procedure adapted for efficient coding of low bit rate voice messaging; and repeating the assigning, comparing, accumulating, mapping, and encoding steps until all data in the second group of pitch and voicing parameters has been coded.
Claim 7
Dependent Claim
126 words
A very low bit rate channel encoding system that codes speech parameters generated from a voice message, the system Comprising: an encoder, comprising: a processor for separating the speech parameters to produce a first group of energy parameters and a second group of pitch and voicing parameters; a non-uniform root-mean-square scalar encoder that encodes and compresses the first group of energy parameters to create a first plurality of encoded data; a three slope vector encoder that compresses, encodes, and combines the second group of pitch and voicing parameters into a single parameter to create a second plurality of encoded data; and a multiplexer that multiplexes the first and second plurality of encoded data create a multiplexed signal for transmission, the multiplexed signal representing the voice message.
Claim 8
Dependent Claim
126 words
A very low bit rate channel encoding system that codes speech parameters generated from a voice message, the system Comprising: an encoder, comprising: a processor for separating the speech parameters to produce a first group of energy parameters and a second group of pitch and voicing parameters; a non-uniform root-mean-square scalar encoder that encodes and compresses the first group of energy parameters to create a first plurality of encoded data; a three slope vector encoder that compresses, encodes, and combines the second group of pitch and voicing parameters into a single parameter to create a second plurality of encoded data; and a multiplexer that multiplexes the first and second plurality of encoded data create a multiplexed signal for transmission, the multiplexed signal representing the voice message.
Claim 9
Dependent Claim
123 words
The very low bit rate channel encoding system according to claim 6 further comprising: a receiver that receives the multiplexed signal, the receiver comprising: a de-multiplexer that demultiplexes the multiplexed signal to create a reproduction of the first and second plurality of encoded data; an inverse non-uniform root-mean-square scalar decoder that decompresses and decodes the first plurality of encoded data; an inverse three slope vector decoder that decompresses and decodes the second plurality of encoded data to recover the second group of pitch and voicing parameters from the single parameter; and a synthesizer that receives the first group of speech parameters and the second group of pitch and voicing parameters and synthesizes a reproduction of said speech parameters to replicate said voice message.
Claim 10
Dependent Claim
71 words
The very low bit rate channel encoding system according to claim 6, wherein the non-uniform root-mean-square scalar encoder operates to perform a discrete cosine transform on the first group of energy parameters resulting in a transformed first group of energy parameters, the non-uniform root-mean-square scalar encoder further comprising: a non-uniform scalar quantizer that operates to quantize the transformed first group of energy parameters to create the first plurality of encoded data.
Claim 11
Dependent Claim
163 words
The very low bit rate channel encoding system according to claim 6, wherein the three slope vector encoder operates to: assign an unvoiced frame to a first pitch value and a voiced frame to a second pitch value; compare a pitch value of a subsequent frame to a pitch value of a prior frame to determine a pitch trend, the pitch trend comprising a constant pitch for a subsequent frame having a like pitch value, an increasing pitch for a subsequent frame having a greater pitch value, and a decreasing pitch for a subsequent frame having a lesser pitch value; accumulate a number of like pitch trend frames until the pitch trend changes; map the pitch trend and the number of like pitch trend frames to a;segment variable; and encode the segment variable using a Huffman encoding procedure adapted for efficient coding of low bit rate voice messaging, until all data in the second group of pitch and voicing parameters has been coded.
References Cited (Backward Citations)
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2000-09-08 | 2001-03-10 | 2001-09-09 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2004-09-08 | 2005-03-10 | 2005-09-09 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2008-09-08 | 2009-03-10 | 2009-09-09 | $7700.00 | Window Closed / Lapsed |