Classifications

• • G—PHYSICS
  • G10—MUSICAL INSTRUMENTS; ACOUSTICS
  • G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
  • G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
  • G10L25/78—Detection of presence or absence of voice signals

Definitions

• the present invention relates to a method for detecting a noisy useful signal.
• the instantaneous amplitude of the signal received or processed can be used as a variable by reference to a threshold determined experimentally.
• thresholds allow a first presumption on the presence or absence of the signal. They are also applicable to any signal. Also, they are supplemented by "confirmation” systems, defining “almost certain” criteria, specific to the type of useful signal, when the nature of this is known a priori.
• Such a complementary system is widely used in speech processing and can consist, for example, in a pitch extraction or in the evaluation of the minimum energy of a vowel.
• the subject of the present invention is a method for detecting a noisy useful signal, determining as rigorously as possible the detection threshold, and which can operate in a self-adaptive manner.
• the expected signal / noise ratio of the signal to be processed there is the expected signal / noise ratio of the signal to be processed, and there is a measurement of the estimated noise alone, measurement digitized on M points, this noise being white or made white, the average energy is calculated.
• noise on these M points we take a slice of N noisy signal points, we calculates the average energy of these N points, the theoretical detection threshold is calculated, the ratio of the two said average energies is calculated, and this ratio is compared with said threshold.
• u (n) s (n) + x (n) n being an integer: 0 ⁇ n ⁇ N-1, s (n) being a useful signal and x (n) a noise.
• y (n) is a measure of the noise x (n) over another time slot free of useful signal.
• the theoretical threshold of 1 is replaced by a threshold ⁇ , calculated as explained below, which takes into account the fact that the signals available are not perfectly ergodic and that U and V are only estimates of the true values of the variances ⁇ u 2 and ⁇ x 2.
• variable U (n) is measured on a time slice
• variable y (n) is measured on another time slice where it is certain that there is no useful signal, but only noise (independent and decorrelated from s (n)).
• the calculation of the density of Z was done by knowing ⁇ s 2 and ⁇ x 2, here the calculation will be done by knowing ⁇ s 2 and ⁇ x 2.
• the density to be calculated will be noted by f z (z: ⁇ 2 s , ⁇ 2 x ).
• U ⁇ s 2 + (1 / N) ⁇ 0 ⁇ n ⁇ N-1 x (n) 2 belongs to ( ⁇ s 2 + ⁇ x 2; (2 / N) ⁇ x 4).
• V belongs to ( ⁇ x 2; (2 / M) ⁇ x 4).
• the activity detection is implemented by using maximum likelihood.
• the probability density of the variable Z is expressed by a function of the form: f k, M (z, r) where r denotes the signal to noise ratio. This probability therefore depends on the signal to noise ratio. Also, the decision rule can only be given with an expected signal-to-noise ratio. Let r o be this expected signal-to-noise ratio.
• the signals s (n), x (n) and y (n) are assumed to be white, Gaussian, centered.
• r o be the expected signal-to-noise ratio
• k M / N. the probability of absence of s (n) is ⁇ o and the probability of presence of s (n) is ⁇ 1.
• the threshold being determined for equality (instead of inequality) between the terms of these two expressions.
• n additive noise
• ⁇ o and ⁇ 1 The probabilities of appearance and absence ( ⁇ o and ⁇ 1) are equal to 0.5.
• V the random variable
• Z For each frame, we calculated Z.
• Decision D 1 when: ln (r + 1) z + k z + k > (M / 4) [z- (r o +1)] 2 - (z-1) 2 z2 + k + ln ⁇ o ⁇ 1
• Decision D 0 when: ln (r + 1) z + k z + k ⁇ (M / 4) [z- (r o +1)] 2 - (z-1) 2 z2 + k + ln ⁇ o ⁇ 1
• a second white Gaussian noise of unit variance was generated, used to calculate V. For each frame, Z was calculated and the above decision rule was applied. We counted the number of errors.
• detection threshold depends on the context.
• a preliminary characterization of noise and speech using measurements based on the maximum likelihood estimation shows that the speech signal to be detected has a signal-to-noise ratio of at least minus 6 dB.
• the processing system uses 128 point signal frames, the sampling frequency being 10 kHz.
• a second threshold is therefore introduced, which makes it possible to decide whether the variable V will be renewed or not.
• This second threshold is chosen at 1.25, which corresponds to noise additive to stationary noise having a signal to noise ratio of -2 dB.
• the processed frame is composed of the same noise as that used as a reference.
• the variable V is replaced by the value of the energy of the processed frame.
• the frame is considered to contain non-stationarity of the noise, and free of speech.
• the frame is considered to be speech.
• the use of two thresholds is generally preferable.
• micro alternation micro opening and closing
• a first pass of the algorithm made it possible to specify the start of the speech.
• a second pass consisted in reading the speech file "upside down", that is to say starting from the microphone closure towards the microphone opening. This then made it possible to specify the end of the speech.
• the same type of application also makes it possible to segment the speech files on which a recognition is carried out.

Applications Claiming Priority (2)

Publications (2)

Family

ID=9413874

Family Applications (1)

Country Status (6)

Cited By (5)

Families Citing this family (22)

Citations (2)

Family Cites Families (9)

• 1991
  • 1991-06-14 FR FR9107323A patent/FR2677828B1/fr not_active Expired - Fee Related
• 1992
  • 1992-06-05 DE DE69225090T patent/DE69225090T2/de not_active Expired - Fee Related
  • 1992-06-05 US US07/972,445 patent/US5337251A/en not_active Expired - Lifetime
  • 1992-06-05 WO PCT/FR1992/000504 patent/WO1992022889A1/fr unknown
  • 1992-06-05 JP JP4511069A patent/JPH06503185A/ja active Pending
  • 1992-06-05 EP EP92401553A patent/EP0518742B1/de not_active Expired - Lifetime

Patent Citations (2)

Non-Patent Citations (2)

Also Published As

Similar Documents

Legal Events