VOICE FILTERING SYSTEM AND METHOD

• US 20190066711A1
• Filed: 11/02/2017
• Published: 02/28/2019
• Est. Priority Date: 08/30/2017
• Status: Abandoned Application

• Current Assignee: Hongfujin Precision Electronics Company Limited (Hon Hai Precision Industry Co., Ltd.), Hon Hai Precision Industry Co., Ltd.
• Original Assignee: Hongfujin Precision Electronics Company Limited (Hon Hai Precision Industry Co., Ltd.), Hon Hai Precision Industry Co., Ltd.
• Inventors: CHUANG, TSUNG-JEN, ZHANG, JUN-WEI, WANG, DAN
• Application Number:
  US15/802,415
• Publication Number:
  US 20190066711A1
• Time in Patent Office: Days
• CPC Class Codes: G10L 2021/02087 the noise being separate speech, e.g. cocktail party
• CPC Class Codes: G10L 2021/02163 Only one microphone
• CPC Class Codes: G10L 21/0208 Noise filtering
• CPC Class Codes: G10L 21/0216 characterised by the method used for estimating noise
• CPC Class Codes: G10L 25/51 for comparison or discrimination
• CPC Class Codes: H04N 23/62 Control of parameters via user interfaces

Abstract

A voice filtering system for controlling an electronic device includes a first storage unit, a voice acquiring unit, a voice analyzing unit, a voice determining unit, and a voice filtering unit. The first storage unit stores at least one first voice and the voice acquiring unit acquires a second voice. The voice analyzing unit analyzes sonic features of the first voice and the second voice and the determining unit detects and determines any differences in the sonic features between the first voice or voices and the second voice. The voice filtering unit filters out any such different features found. A voice filtering method is also presented.

Claims

1. A voice filtering system comprising:

a first storage unit configured to store at least one first voice;

a voice acquiring unit configured to acquire a second voice;

a voice analyzing unit configured to analyze features of the first voice and the second voice;

a determining unit configured to determine whether there is a voice different from the feature of the first voice in the second voice;

a voice filtering unit configured to filter the voice different from the feature of the first voice in the second voice; and

at least one processor for executing programs associated with at least one of the first storage unit, the voice acquiring unit, the voice analyzing unit, the determining unit and the voice filtering unit.

2. The voice filtering system as claimed in claim 1, wherein the feature comprises timbre, tone and loudness.

3. The voice filtering system as claimed in claim 1, wherein the voice acquiring unit comprises a microphone and the microphone captures the second voice.

4. The voice filtering system as claimed in claim 1, further comprising a second storage unit, wherein the second storage unit stores the filtered second voice.

5. The voice filtering system as claimed in claim 1, further comprising:

a switch unit, wherein the switch unit controls the voice filtering unit to turn on or off.

6. The voice filtering system as claimed in claim 1, further comprising:

an image capturing unit configured to capture images; and

a video generating unit configured to generate a video by the image and the second voice.

7. The voice filtering system as claimed in claim 1, further comprising:

a selecting unit configured to select at least one target person on the image, wherein the first voice is a voice of the target person.

8. The voice filtering system as claimed in claim 7, wherein the selecting unit comprises a touch sensing unit and a people determining unit, the touch sensing unit senses a touch position on the image, when there is no touch in a preset time after sensing a touch, the people determining unit determines an image part corresponding to the touch position is the target person.

9. The voice filtering system as claimed in claim 8, wherein the voice filtering system further comprises a labeling unit, the selecting unit selects one target person, when the second voice is a voice of the target person, the labeling unit labels the target person on the image with a first mark.

10. The voice filtering system as claimed in claim 9, wherein the first mark is a twinkling box.

11. A voice filtering method comprising:

storing at least one first voice;

acquiring a second voice;

analyzing features of the first voice and the second voice;

determining whether there is a voice different from the feature of the first voice in the second voice; and

filtering the voice different from the feature of the first voice in the second voice.

12. The voice filtering method as claimed in claim 11, wherein the feature comprises timbre, tone and loudness.

13. The voice filtering method as claimed in claim 11, further comprising acquiring the second voice by a microphone.

14. The voice filtering method as claimed in claim 11, further comprising:

storing the filtered second voice.

15. The voice filtering method as claimed in claim 11, further comprising:

activating a voice filtering function; and

wherein when the voice filtering function is activated, the voice different from the feature of the first voice in the second voice is able to be filtered.

16. The voice filtering method as claimed in claim 11, further comprising:

capturing images; and

generating a video by the image and the second voice.

17. The voice filtering method as claimed in claim 11, further comprising:

selecting at least one target person on the image, wherein the first voice is a voice of the target person.

18. The voice filtering method as claimed in claim 17, wherein selecting at least one target person on the image comprises:

sensing a touch position on the image; and

when there is no touch in a preset time after sensing a touch, determining an image part corresponding to the touch position is the target person.

19. The voice filtering method as claimed in claim 18, further comprising selecting one target person, and further comprising:

when the second voice is a voice of the target person, labeling the target person on the image with a first mark.

20. The voice filtering method as claimed in claim 11, wherein the first mark is a twinkling box.