EP1860645A2 - Segmentation automatique dans la synthèse vocale - Google Patents

Segmentation automatique dans la synthèse vocale Download PDF

Info

Publication number
EP1860645A2
EP1860645A2 EP07116265A EP07116265A EP1860645A2 EP 1860645 A2 EP1860645 A2 EP 1860645A2 EP 07116265 A EP07116265 A EP 07116265A EP 07116265 A EP07116265 A EP 07116265A EP 1860645 A2 EP1860645 A2 EP 1860645A2
Authority
EP
European Patent Office
Prior art keywords
hmms
phone
labels
spectral
phone labels
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
EP07116265A
Other languages
German (de)
English (en)
Other versions
EP1860645A3 (fr
Inventor
Alistair D. Conkie
Yeon-Jun Kim
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
AT&T Corp
Original Assignee
AT&T Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from US10/341,869 external-priority patent/US7266497B2/en
Application filed by AT&T Corp filed Critical AT&T Corp
Publication of EP1860645A2 publication Critical patent/EP1860645A2/fr
Publication of EP1860645A3 publication Critical patent/EP1860645A3/fr
Withdrawn legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L13/00—Speech synthesis; Text to speech systems
    • G10L13/06—Elementary speech units used in speech synthesisers; Concatenation rules

Definitions

  • the present invention relates to systems and methods for automatic segmentation in speech synthesis. More particularly, the present invention relates to systems and methods for automatic segmentation in speech synthesis by combining a Hidden Markov Model (HMM) approach with spectral boundary correction.
  • HMM Hidden Markov Model
  • TTS text-to-speech
  • ASR automatic speech recognition
  • the quality of a TTS system is often dependent on the speech inventory and on the accuracy with which the speech inventory is segmented and labeled.
  • the speech or acoustic inventory usually stores speech units (phones, diphones, half-phones, etc.) and during speech synthesis, units are selected and concatenated to create the synthetic speech.
  • the speech inventory should be accurately segmented and labeled in order to avoid noticeable errors in the synthetic speech.
  • Automatic segmentation of a speech inventory plays an important role in significantly reducing reduce the human effort that would otherwise be require to build, train, and/or segment speech inventories. Automatic segmentation is particularly useful as the amount of speech to be processed becomes larger.
  • HMM Hidden Markov Model
  • hand-labeled bootstrapping may require a month of labeling by a phonetic expert to prepare training data for speaker-dependent HMMs (SD HMMs).
  • SD HMMs speaker-dependent HMMs
  • SI HMMs speaker-independent HMMs
  • An HMM-based approach is somewhat limited in its ability to remove discontinuities at concatenation points because the Viterbi alignment used in an HMM-based approach tries to find the best HMM sequence when given a phone transcription and a sequence of HMM parameters rather than the optimal boundaries between adjacent units or phones.
  • an HMM-based automatic segmentation system may locate a phone boundary at a different position than expected, which results in mismatches at unit concatenation points and in speech discontinuities. There is therefore a need to improve automatic segmentation.
  • the present invention overcomes these and other limitations and relates to systems and methods for automatically segmenting a speech inventory. More particularly, the present invention relates to systems and methods for automatically segmenting phones and more particularly to automatically segmenting a speech inventory by combining an HMM-based approach with spectral boundary correction.
  • automatic segmentation begins by bootstrapping a set of HMMs with speaker-independent HMMs.
  • the set of HMMs is initialized, re-estimated, and aligned to produce the labeled units or phones.
  • the boundaries of the phone or unit labels that result from the automatic segmentation are corrected using spectral boundary correction.
  • the resulting phones are then used as seed data for HMM initialization and re-estimation. This process is performed iteratively.
  • a phone boundary is defined, in one embodiment, as the position where the maximal concatenation cost concerning spectral distortion is located.
  • Euclidean distance between mel frequency cepstral coefficients (MFCCs) is often used to calculate spectral distortions
  • the present invention utilizes a weighted slop metric.
  • the bending point of a spectral transition often coincides with a phone boundary.
  • the spectral-boundary-corrected phones are then used to initialize, re-estimate and align the HMMs iteratively.
  • the labels that have been re-aligned using spectral boundary correction are used as feedback for iteratively training the HMMs. In this manner, misalignments between target phone boundaries and boundaries assigned by automatic segmentation can be reduced.
  • Speech inventories are used, for example, in text-to-speech (TTS) systems and in automatic speech recognition (ASR) systems.
  • the quality of the speech that is rendered by concatenating the units of the speech inventory represents how well the units or phones are segmented.
  • the present invention relates to systems and methods for automatically segmenting speech inventories and more particularly to automatically segmenting a speech inventory by combining an HMM-based segmentation approach with spectral boundary correction. By combining an HMM-based segmentation approach with spectral boundary correction, the segmental quality of synthetic speech in unit-concatenative speech synthesis is improved.
  • An exemplary HMM-based approach to automatic segmentation usually includes two phases: training the HMMs, and unit segmentation using the Viterbi alignment.
  • each phone or unit is defined as an HMM prior to unit segmentation and then trained with a given phonetic transcription and its corresponding feature vector sequence.
  • TTS systems often require more accuracy in segmentation and labeling than do ASR systems.
  • FIG 1 illustrates an exemplary TTS system that converts text to speech.
  • the TTS system 100 converts the text 110 to audible speech 118 by first performing a linguistic analysis 112 on the text 110.
  • the linguistic analysis 112 includes, for example, applying weighted finite state transducers to the text 110.
  • each segment is associated with various characteristics such as segment duration, syllable stress, accent status, and the like.
  • Speech synthesis 116 generates the synthetic speech 118 by concatenating segments of natural speech from a speech inventory 120.
  • the speech inventory 120 in one embodiment, usually includes a speech waveform and phone labeled data.
  • the boundary of a unit for segmentation purposes is defined as being where one unit ends and another unit begins.
  • the segmentation must occur as close to the actual unit boundary as possible. This boundary often naturally occurs within a certain time window depending on the class of the two adjacent units. In one embodiment of the present invention, only the boundaries within these time windows are examined during spectral boundary correction in order to obtain more accurate unit boundaries. This prevents a spurious boundary from being inadvertently recognized as the phone boundary, which would lead to discontinuities in the synthetic speech.
  • Figure 2 illustrates an exemplary method for automatically segmenting phones or units and illustrates three examples of seed data to begin the initialization of a set of HMMs.
  • Seed data can be obtained using, for example: hand-labeled bootstrap 202, speaker-independent (SI) HMM bootstrap 204, and a flat start 206.
  • Hand-labeled bootstrapping which utilizes a specific speaker's hand-labeled speech data, results in the most accurate HMM modeling and is often called speaker-dependent HMM (SD HMM). While SD HMMs are generally used for automatic segmentation in speech synthesis, they have the disadvantage of being quite time-consuming to prepare.
  • One advantage of the present invention is to reduce the amount of time required to segment the speech inventory.
  • SI HMMs for American English trained with the TIMIT speech corpus, were used in the preparation of seed phone labels. With the resulting labels, SD HMMs for an American male speaker were trained to provide the segmentation for building an inventory of synthesis units.
  • One advantage of bootstrapping with SI HMMs is that all of the available speech data can be used as training data if necessary.
  • the automatic segmentation system includes ARPA phone HMMs that use three-state left-to-right models with multiple mixture of Gaussian density.
  • standard HMM input parameters which include twelve MFCCs (Mel frequency cepstral coefficients), normalized energy, and their first and second order delta coefficients, are utilized.
  • the SD HMMs bootstrapped with SI HMMs result in phones being labeled with an accuracy of 87.3% ( ⁇ 20 ms, compared to hand labeling).
  • Many errors are caused by differences between the speaker's actual pronunciations and the given pronunciation lexicon, i.e., errors by the speaker or the lexicon or effects of spoken language such as contractions. Therefore, speaker-individual pronunciation variations have to be added to the lexicon.
  • Figure 2 illustrates a flow diagram for automatic segmentation that combines an HMM-based approach with iterative training and spectral boundary correction.
  • Initialization 208 occurs using the data from the hand-labeled bootstrap 202, the SI HMM bootstrap 204, or from a flat start 206. After the HMMs are initialized, the HMMs are re-estimated (210). Next, embedded re-estimation 212 is performed. These actions - initialization 208, re-estimation 210, and embedded re-estimation 212 - are an example of how HMMs are trained from the seed data.
  • a Viterbi alignment 214 is applied to the HMMs in one embodiment to produce the phone labels 216.
  • the phones are labeled and can be used for speech synthesis.
  • spectral boundary correction is applied to the resulting phone labels 216.
  • the resulting phones are trained and aligned iteratively. In other words, the phone labels that have been re-aligned using spectral boundary correction are used as input to initialization 208 iteratively.
  • the hand-labeled bootstrapping 202, SI HMM bootstrapping 204, and the flat start 206 are usually used the first time the HMMs are trained. Successive iterations use the phone labels that have been aligned using spectral boundary correction 218.
  • a reduction of mismatches between phone boundary labels is expected when the temporal alignment of the feed-back labeling is corrected.
  • Phone boundary corrections can be done manually or by rule-based approaches. Assuming that the phone labels assigned by an HMM-based approach are relatively accurate, automatic phone boundary correction concerning spectral features improves the accuracy of the automatic segmentation.
  • One advantage of the present invention is to reduce or minimize the audible signal discontinuities caused by spectral mismatches between two successive concatenated units.
  • a phone boundary can be defined as the position where the maximal concatenation cost concerning spectral distortion, i.e., the spectral boundary, is located.
  • the Euclidean distance between MFCCs is most widely used to calculate spectral distortions.
  • the present embodiment uses instead the weighted slope metric (see Equation (1) below).
  • S L and S R are 256 point FFTs (fast Fourier transforms) divided into K critical bands.
  • the S L and S R vectors represent the spectrum to the left and the right of the boundary, respectively.
  • E S L and E S R are spectral energy
  • ⁇ S L ( i ) and ⁇ S R ( i ) are the i th critical band spectral slopes of S L and S R (see Figure 3)
  • u E , u(i) are weighting factors for the spectral energy difference and the i th spectral transition.
  • Spectral transitions play an important role in human speech perception.
  • Figure 3 which illustrates adjacent spectral slopes, more fully illustrates the bending point of a spectral transition.
  • the spectral slope 304 corresponds to the i th critical band of S L
  • the spectral slope 306 corresponds to the i th critical band of S R .
  • the bending point 302 of the spectral transition usually coincides with a phone boundary. Using spectral boundaries identified in this fashion, spectral boundary correction 218 can be applied to the phone labels 216, as illustrated in Figure 2.
  • E S L - E S R which is the absolute energy difference in Equation (1), is modified to distinguish K critical bands, as in Equation (2):
  • the automatic detector described above may produce a number of spurious peaks.
  • a context-dependent time window in which the optimal phone boundary is more likely to be found is used. The phone boundary is checked only within the specified context-dependent time window.
  • Temporal misalignment tends to vary in time depending on the contexts of two adjacent phones. Therefore, the time window for finding the local maximum of spectral boundary distortion is empirically determined, in this embodiment, by the adjacent phones as illustrated in the following table.
  • This table represents context-dependent time windows (in ms) for spectral boundary correction (V: Vowel, P: Unvoiced stop, B: Voiced stop, S: Unvoiced fricative, Z: Voiced fricative, L: Liquid, N: Nasal).
  • BOUNDARY Time window (ms) BOUNDARY Time window (ms) V-V -4.5 ⁇ 50 P-V -1.6 ⁇ 30 V-N -4.8 ⁇ 30 N-V 0 ⁇ 30 V-B -13.9 ⁇ 30 B-V 0 ⁇ 20 V-L -23.2 ⁇ 40 L-V 11.1 ⁇ 30 V-P 2.2 ⁇ 20 S-V 2.7 ⁇ 20 V-Z -15.8 ⁇ 30 Z-V 15.4 ⁇ 40
  • the present invention relates to a method for automatically segmenting phones or other units by combining HMM-based segmentation with spectral features using spectral boundary correction. Misalignments between target phone boundaries and boundaries assigned by automatic segmentation are reduced and result in more natural synthetic speech. In other words, the concatenation points are less noticeable and the quality of the synthetic speech is improved.
  • the embodiments of the present invention may comprise a special purpose or general purpose computer including various computer hardware, as discussed in greater detail below.
  • Embodiments within the scope of the present invention may also include computer-readable media for carrying or having computer-executable instructions or data structures stored thereon.
  • Such computer-readable media can be any available media that can be accessed by a general purpose or special purpose computer.
  • Such computer-readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to carry or store desired program code means in the form of computer-executable instructions or data structures and which can be accessed by a general purpose or special purpose computer.
  • Computer-executable instructions include, for example, instructions and data which cause a general purpose computer, special purpose computer, or special purpose processing device to perform a certain function or group of functions.
  • Computer-executable instructions also include program modules which are executed by computers in stand alone or network environments.
  • program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types.
  • Computer-executable instructions, associated data structures, and program modules represent examples of the program code means for executing steps of the methods disclosed herein. The particular sequence of such executable instructions or associated data structures represents examples of corresponding acts for implementing the functions described in such steps.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Telephonic Communication Services (AREA)
  • Image Analysis (AREA)
  • Telephone Function (AREA)
  • Machine Translation (AREA)
EP07116265A 2002-03-29 2003-03-27 Segmentation automatique dans la synthèse vocale Withdrawn EP1860645A3 (fr)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
US36904302P 2002-03-29 2002-03-29
US10/341,869 US7266497B2 (en) 2002-03-29 2003-01-14 Automatic segmentation in speech synthesis
EP03100795A EP1394769B1 (fr) 2002-03-29 2003-03-27 Segmentation automatique en synthèse de parole

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
EP03100795A Division EP1394769B1 (fr) 2002-03-29 2003-03-27 Segmentation automatique en synthèse de parole

Publications (2)

Publication Number Publication Date
EP1860645A2 true EP1860645A2 (fr) 2007-11-28
EP1860645A3 EP1860645A3 (fr) 2008-09-03

Family

ID=38621945

Family Applications (2)

Application Number Title Priority Date Filing Date
EP07116265A Withdrawn EP1860645A3 (fr) 2002-03-29 2003-03-27 Segmentation automatique dans la synthèse vocale
EP07116266A Withdrawn EP1860646A3 (fr) 2002-03-29 2003-03-27 Segmentation automatique dans la synthèse vocale

Family Applications After (1)

Application Number Title Priority Date Filing Date
EP07116266A Withdrawn EP1860646A3 (fr) 2002-03-29 2003-03-27 Segmentation automatique dans la synthèse vocale

Country Status (1)

Country Link
EP (2) EP1860645A3 (fr)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1035537A2 (fr) * 1999-03-09 2000-09-13 Matsushita Electric Industrial Co., Ltd. Identification de régions de recouvrement d'unités pour un système de synthèse de parole par concaténation

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1035537A2 (fr) * 1999-03-09 2000-09-13 Matsushita Electric Industrial Co., Ltd. Identification de régions de recouvrement d'unités pour un système de synthèse de parole par concaténation

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
BRUGNARA F ET AL: "AUTOMATIC SEGMENTATION AND LABELING OF SPEECH BASED ON HIDDEN MARKOV MODELS" SPEECH COMMUNICATION, ELSEVIER SCIENCE PUBLISHERS, AMSTERDAM, NL, vol. 12, no. 4, 1 August 1993 (1993-08-01), pages 357-370, XP000393652 ISSN: 0167-6393 *
HON H ET AL: "Automatic generation of synthesis units for trainable text-to-speech systems" ACOUSTICS, SPEECH AND SIGNAL PROCESSING, 1998. PROCEEDINGS OF THE 1998 IEEE INTERNATIONAL CONFERENCE ON SEATTLE, WA, USA 12-15 MAY 1998, NEW YORK, NY, USA,IEEE, US, 12 May 1998 (1998-05-12), pages 293-296, XP010279159 ISBN: 0-7803-4428-6 *
TOLEDANO D T: "Neural network boundary refining for automatic speech segmentation" 2000 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL, vol. 6, 5 June 2000 (2000-06-05), pages 3438-3441, XP010505636 *

Also Published As

Publication number Publication date
EP1860645A3 (fr) 2008-09-03
EP1860646A2 (fr) 2007-11-28
EP1860646A3 (fr) 2008-09-03

Similar Documents

Publication Publication Date Title
US8131547B2 (en) Automatic segmentation in speech synthesis
Kim et al. Automatic segmentation combining an HMM-based approach and spectral boundary correction.
DiCanio et al. Using automatic alignment to analyze endangered language data: Testing the viability of untrained alignment
US7856357B2 (en) Speech synthesis method, speech synthesis system, and speech synthesis program
Ljolje et al. Automatic speech segmentation for concatenative inventory selection
Arslan Speaker transformation algorithm using segmental codebooks (STASC)
US20060259303A1 (en) Systems and methods for pitch smoothing for text-to-speech synthesis
Malfrère et al. Phonetic alignment: speech synthesis-based vs. Viterbi-based
US20040030555A1 (en) System and method for concatenating acoustic contours for speech synthesis
Hirai et al. Using 5 ms segments in concatenative speech synthesis
US20030195743A1 (en) Method of speech segment selection for concatenative synthesis based on prosody-aligned distance measure
US20060074678A1 (en) Prosody generation for text-to-speech synthesis based on micro-prosodic data
Hoffmann et al. Fully automatic segmentation for prosodic speech corpora.
Toledano et al. Trying to mimic human segmentation of speech using HMM and fuzzy logic post-correction rules
Blackburn et al. Towards improved speech recognition using a speech production model.
Chou et al. Corpus-based Mandarin speech synthesis with contextual syllabic units based on phonetic properties
Gonzalvo Fructuoso et al. Linguistic and mixed excitation improvements on a HMM-based speech synthesis for Castilian Spanish
Matoušek et al. Experiments with automatic segmentation for Czech speech synthesis
EP1860645A2 (fr) Segmentation automatique dans la synthèse vocale
Mustafa et al. Developing an HMM-based speech synthesis system for Malay: a comparison of iterative and isolated unit training
Youssef et al. An Arabic TTS system based on the IBM trainable speech synthesizer
Carvalho et al. Concatenative speech synthesis for European Portuguese
Jafri et al. Statistical formant speech synthesis for Arabic
Rouibia et al. Unit selection for speech synthesis based on a new acoustic target cost.
Carvalho et al. Automatic segment alignment for concatenative speech synthesis in portuguese

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

AC Divisional application: reference to earlier application

Ref document number: 1394769

Country of ref document: EP

Kind code of ref document: P

AK Designated contracting states

Kind code of ref document: A2

Designated state(s): DE FI FR GB NL

PUAL Search report despatched

Free format text: ORIGINAL CODE: 0009013

AK Designated contracting states

Kind code of ref document: A3

Designated state(s): DE FI FR GB NL

RIC1 Information provided on ipc code assigned before grant

Ipc: G10L 15/14 20060101ALI20080725BHEP

Ipc: G10L 13/06 20060101AFI20071024BHEP

17P Request for examination filed

Effective date: 20090302

17Q First examination report despatched

Effective date: 20090403

AKX Designation fees paid

Designated state(s): DE FI FR GB NL

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE APPLICATION IS DEEMED TO BE WITHDRAWN

18D Application deemed to be withdrawn

Effective date: 20090825