Voice synthesis device, voice synthesis method, bone conduction helmet and hearing aid
AssigneeBoe Technology Group Co., Ltd.
InventorXuan Liang
Life of this patent
Estimating the term…
Abstract
The present disclosure provides a voice synthesis device, a voice synthesis method, a bone conduction helmet and a hearing aid. The voice synthesis device of the present disclosure comprises an aggregation node and a plurality of detection nodes, wherein the detection nodes are used for detecting external voice signals and transmitting the external voice signals to the aggregation node. The aggregation node is used for filtering the detected voice signals and synthesizing the voice signals sent by trusted detection nodes. Thus, the bit error rate of voice signals can be reduced significantly, and the accuracy of voice signals can be improved.
Claims16
A voice synthesis device comprising:
- a plurality of detection nodes configured to detect voice signals, and
- an aggregation node configured to receive the voice signals sent by the plurality of detection nodes, to determine whether each of the plurality of detection nodes is trusted, and to synthesize the voice signals sent by the detection nodes which are determined to be trusted to obtain a synthesized voice signal,
- wherein the aggregation node comprises:
- a first wireless transceiver configured to receive the voice signals sent by the plurality of detection nodes;
- and
- a first program controller configured to determine whether each of the plurality of detection nodes is trusted, and to synthesize the voice signals sent by the detection nodes which are determined to be trusted to obtain the synthesized voice signal,
- wherein each detection node of the plurality of detection nodes is further configured to send a feedback signal to the aggregation node in response to receiving a broadcast signal sent by the aggregation node, and
- wherein the aggregation node is further configured to send the broadcast signal to the plurality of detection nodes, to determine a trust degree for each of the detection nodes in response to receiving the feedback signal from each of the plurality of detection nodes, and to determine whether each detection node is trusted based on its trust degree;
- wherein the trust degree for each detection node is determined as:
- K
- i
-
- (
- t
- )
- =
-
- Y
- i
-
- (
- t
- )
- -
- 1
- /
- n
-
- ∑
- i
- =
- 1
- n
-
-
- Y
- i
-
- (
- t
- )
-
- 1
- /
- n
-
- ∑
- i
- =
- 1
- n
-
-
- Y
- i
-
- (
- t
- )
- ,
- wherein Ki(t) denotes the trust degree for a detection node i at time t, n denotes the number of the plurality of detection nodes, and Yi(t) denotes a voice signal detected by the detection node i at time t, as the feedback signal.
A voice synthesis method, comprising:
- receiving voice signals sent by a plurality of detection nodes,
- determining whether each of the plurality of detection nodes is trusted, and
- synthesizing the voice signals sent by the detection nodes which are determined to be trusted to obtain a synthesized voice signal;
- wherein determining whether each of the plurality of detection nodes is trusted comprises:
- sending a broadcast signal to the plurality of detection nodes;
- determining a trust degree for each of the detection nodes in response to receiving a feedback signal from each of the plurality of detection nodes;
- and
- determining whether each detection node is trusted based on its trust degree,
- wherein the trust degree for each detection node is determined as:
- K
- i
-
- (
- t
- )
- =
-
- Y
- i
-
- (
- t
- )
- -
- 1
- /
- n
-
- ∑
- i
- =
- 1
- n
-
- Y
- i
-
- (
- t
- )
-
- 1
- /
- n
-
- ∑
- i
- =
- 1
- n
-
- Y
- i
-
- (
- t
- )
- ,
- wherein Ki(t) denotes the trust degree for a detection node i at time t, n denotes the number of the plurality of detection nodes, and Yi(t) denotes a voice signal detected by the detection node i at time t, as the feedback signal.
Dependencies are inferred from the claim text.