Back to results
US 6,470,309 B1Granted
Subframe-based correlation
AssigneeTexas Instruments Incorporated
InventorAlan V. McCree
Life of this patent
Estimating the term…
Abstract
A subframe-based correlation method for pitch and voicing is provided by finding the pitch track through a speech frame that minimizes pitch prediction residual energy over the frame. The method scans the range of possible time lags T and computes for each subframe within a given range of T the maximum correlation value and further finds the set of subframe lags to maximize the correlation over all of possible pitch lags.
Claims25
A subframe-based correlation method comprising the steps of:
- varying lag times T over all pitch range in a speech frame;
- determining pitch lags for each subframe within said overall range that maximize the correlation value according to ∑ n ( x n x n - T s ) 2 ∑ n x n - T 2
- provided the pitch lags across the subframe are within a given constrained range, where Ts is the subframe lag, xn is the nth sample of the input signal and the Σn includes all samples in subframes.
A subframe-based correlation method comprising the steps of:
- varying lag times T over all pitch range in a speech frame;
- determining pitch lags for each subframe within said overall range that maximize the correlation value according to ∑ n ( x n x n - T s ) 2 ∑ n x n - T 2 × w ( T s )
- provided the pitch lags across the subframe are within a given constrained range, where Ts is the subframe lag, xn is the nth sample of the input signal w(Ts) is a weighting function to penalize pitch doubles and the Σn includes all samples in subframes.
A method of determining normalized correlation coefficient comprising the steps of:
- providing a set of subframe lags Ts and computing the normalized correlation for that set of Ts according to ρ ( T ) = ∑ s = 1 N s ( ∑ n x n x n - T s ) 2 ∑ n x n - T s 2 ∑ s = 1 N s ∑ n x n 2
- where Ns is the number of samples in a frame and xn is the nth sample.
A subframe-based correlation method comprising the steps of:
- varying lag times T over all pitch range in a speech frame;
- determining pitch lags for each subframe within said overall range that maximize the correlation value according to max { T s } [ ∑ s = 1 N s 2 [ ( ∑ n x n x n + T s ) 2 ∑ n x n + T s 2 × w ( T s ) ] + ∑ s = N s 2 + 1 N s [ ( ∑ n x n x n - T s ) 2 ∑ n x n - T s 2 × w ( T s ) ] ]
- provided the pitch lags across the subframe are within a given constrained range, where Ts is the subframe lag, xn is the nth sample of the input signal, Ns is samples in a frame, w(Ts) is a weighting function for doubles and the Σn includes all samples in subframes.
A voice coder comprising:
- an encoder for voice input signals, said encoder including
- a pitch estimator for determining pitch of said input signals;
- a synthesizer coupled to said encoder and responsive to said input signals for providing synthesized voice output signals, said synthesizer coupled to said pitch estimator for providing synthesized output based for said determined pitch of said input signals;
- said pitch estimator determining pitch according to:
- T = max T = lower upper [ ∑ s = 1 N s max T s = T - Δ T + Δ [ ( ∑ n x n x n - T s ) 2 ∑ n x n - T 2 ] ]
- where Ts is the subframe lag, xn is the nth sample of the input signal, ρn, includes all samples in the subframe, T is determining maximum correlation values of subframes for each value T, Ns is the number of samples in a frame and Δ is the constrained range of the subframe.
3 more elements
A voice coder comprising:
- an encoder for voice input signals, said encoder including means for determining sets of subframe lags Ts over a pitch range;
- and
- means for determining a normalized correlation coefficient ρ(T) for a pitch path in each frequency band where ρ(T) is determined by ρ ( T ) = ∑ s = 1 N s ( ∑ n x n x n - T s ) 2 ∑ n x n - T s 2 ∑ s = 1 N s ∑ n x n s
- where Ns is the number of samples in a frame, and xn is the nth sample.
1 more element
A voice coder comprising:
- an encoder for voice input signals said encoder including
- a pitch estimator for determining pitch of said input signals;
- a synthesizer coupled to said encoder and responsive to said input signals for providing synthesized voice output signals, said synthesizer coupled to said pitch estimator for providing synthesized output based for said determined pitch of said input signals;
- said pitch estimator determining pitch according to:
- T = [ ( ∑ n x n x n - T s ) 2 ∑ n x n - T 2 ]
- where Ts is the subframe lag, xn is the nth sample of the input signal and Σn includes all samples in subframes.
3 more elements
A method of determining normalized correlation coefficient at fractional pitch period comprising the steps of:
- providing a set of subframe lags Ts;
- finding a fraction q by c ( 0 , T s + 1 ) c ( T s , T s ) - c ( 0 , T s ) c ( T s , T s + 1 ) c ( 0 , T s + 1 ) [ c ( T s , T s ) - c ( T s , T s + 1 ) ] + c ( 0 , T s ) [ c ( T s + 1 , T s + 1 ) - c ( T s , T s + 1 ) ]
- where c is the inner product of two vectors and the normalized correlation for subframe is determined by;
- ρ s ( T s + q ) = ( 1 - q ) c ( 0 , T s ) + qc ( 0 , T s + 1 ) c ( 0 , 0 ) [ ( 1 - q ) 2 ( T s , T s ) + 2 q ( 1 - q ) c ( T s , T s + 1 ) + q 2 c ( T s + 1 , T s + 1 ) ] ;
- and substituting ρs(Ts+q) for ρs in ρ ( T ) = ∑ s = 1 N s p s ρ s 2 ( T s ) ∑ s = 1 N s p s where p s = ∑ n x n 2 .
2 more elements
Dependencies are inferred from the claim text.