TW201614641A - Method and apparatus for speech enhancement based on source separation - Google Patents
Method and apparatus for speech enhancement based on source separationInfo
- Publication number
- TW201614641A TW201614641A TW104128192A TW104128192A TW201614641A TW 201614641 A TW201614641 A TW 201614641A TW 104128192 A TW104128192 A TW 104128192A TW 104128192 A TW104128192 A TW 104128192A TW 201614641 A TW201614641 A TW 201614641A
- Authority
- TW
- Taiwan
- Prior art keywords
- speech
- sparsity
- noise
- activations
- spectral model
- Prior art date
Links
- 238000000926 separation method Methods 0.000 title abstract 2
- 238000000034 method Methods 0.000 title 1
- 230000004913 activation Effects 0.000 abstract 5
- 238000001994 activation Methods 0.000 abstract 5
- 230000003595 spectral effect Effects 0.000 abstract 5
- 239000011159 matrix material Substances 0.000 abstract 1
- 238000005457 optimization Methods 0.000 abstract 1
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
- G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0272—Voice signal separating
-
- 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
- G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
- G10L21/0216—Noise filtering characterised by the method used for estimating noise
- G10L2021/02161—Number of inputs available containing the signal or the noise to be suppressed
- G10L2021/02163—Only one microphone
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Quality & Reliability (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Circuit For Audible Band Transducer (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present embodiments provide speech enhancement based on source separation techniques. Specifically, we use a universal spectral model for speech, and train the spectral model for noise and activations for speech/noise based on the universal spectral model for speech and input noisy speech. We formulate the optimization problem using a cost function that includes a divergence function and a sparsity penalty function, wherein the penalty function is based on the notion of relative group sparsity. The sparsity penalty function includes two parts: a sparsity-promoting part for the groups (activations for some groups become zero) and an anti-sparsity-promoting part for the whole activation matrix corresponding to the speech model (i.e., the activations for speech as a whole does not become zero). Based on the universal spectral model for speech, the spectral model for noise, and activations for speech/noise, we can estimate the speech/noise included in the input noisy speech.
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP14306540 | 2014-09-30 |
Publications (1)
Publication Number | Publication Date |
---|---|
TW201614641A true TW201614641A (en) | 2016-04-16 |
Family
ID=51730467
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW104128192A TW201614641A (en) | 2014-09-30 | 2015-08-27 | Method and apparatus for speech enhancement based on source separation |
Country Status (2)
Country | Link |
---|---|
TW (1) | TW201614641A (en) |
WO (1) | WO2016050725A1 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113656747A (en) * | 2021-08-13 | 2021-11-16 | 南京理工大学 | Array self-adaptive beam forming method under multiple expected signals based on branch and bound |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108076238A (en) * | 2016-11-16 | 2018-05-25 | 艾丽西亚(天津)文化交流有限公司 | A kind of science and technology service packet audio mixing communicator |
CN108573698B (en) * | 2017-03-09 | 2021-06-08 | 中国科学院声学研究所 | Voice noise reduction method based on gender fusion information |
CN109346097B (en) * | 2018-03-30 | 2023-07-14 | 上海大学 | Speech enhancement method based on Kullback-Leibler difference |
US11227621B2 (en) | 2018-09-17 | 2022-01-18 | Dolby International Ab | Separating desired audio content from undesired content |
CN111710343B (en) * | 2020-06-03 | 2022-09-30 | 中国科学技术大学 | Single-channel voice separation method on double transform domains |
CN113823316B (en) * | 2021-09-26 | 2023-09-12 | 南京大学 | Voice signal separation method for sound source close to position |
-
2015
- 2015-08-27 TW TW104128192A patent/TW201614641A/en unknown
- 2015-09-29 WO PCT/EP2015/072344 patent/WO2016050725A1/en active Application Filing
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113656747A (en) * | 2021-08-13 | 2021-11-16 | 南京理工大学 | Array self-adaptive beam forming method under multiple expected signals based on branch and bound |
Also Published As
Publication number | Publication date |
---|---|
WO2016050725A1 (en) | 2016-04-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
TW201614641A (en) | Method and apparatus for speech enhancement based on source separation | |
MX2017002593A (en) | Event stream transformations. | |
GB201208373D0 (en) | Mechanism for synchronising devices,system and method | |
CO2017007028A2 (en) | Headless completion of tasks within personal digital assistants | |
WO2014102548A3 (en) | Search system and corresponding method | |
GB2566420A (en) | Providing debug information on production containers using debug containers | |
MX2016013015A (en) | Methods and systems of handling a dialog with a robot. | |
MX2015003552A (en) | Emergency vehicle maneuver communications. | |
MX2016014234A (en) | System and method for the creation and use of visually-diverse high-quality dynamic layouts. | |
MX2016006034A (en) | Determining vehicle occupant location. | |
TWD179112S (en) | Model vehicle platform | |
MX2017004763A (en) | Use of cannabidiol in the treatment of tuberous sclerosis complex. | |
MX2018003823A (en) | Information presenting device and information presenting method. | |
TWD170361S (en) | All-terrain vehicle | |
TWD167955S (en) | Portion of biosensor | |
GB201305379D0 (en) | Methods and systems for enrolling biometric data | |
PH12015000372A1 (en) | Conversion of documents of different types to a uniform and an editable or a searchable format | |
GB2522579A (en) | Computing device with force-triggered non-visual responses | |
WO2014121234A3 (en) | Method and apparatus for contextual text to speech conversion | |
TWD180195S (en) | Portion of a bottle | |
TWD199437S (en) | Hammer | |
EA201791047A1 (en) | SYSTEM AND REGULATORY METHOD FOR SYSTEMS AND METHODS FOR POWER GENERATION | |
WO2014152179A3 (en) | Pet insurance system and method | |
MX2015014413A (en) | Acoustic impulse response simulation. | |
TWD196188S (en) | Hammer |