30results about How to "Good voice effect" patented technology

The invention discloses a method and a device for handling receiving voice, which belongs to the technical field of communication. The method comprises the steps of: acquiring the use condition information of a voice output device; respectively regulating the signal intensity of the receiving voice received from each frequency band according to the use condition information of the voice output device, and outputting the regulated receiving voice through the voice output device, wherein the frequency bands are obtained by dividing the voice bandwidth of the voice output device. The device comprises an acquisition module and a handling module. In the invention, by acquiring and in accordance with the use condition information of the voice output device, the signal intensities of the receiving voice received from each frequency band are regulated respectively, the signal intensity of voice in a frequency band with higher signal intensity attenuation is intensified greatly, the signal intensity of voice in a frequency band with lower signal intensity attenuation is intensified slightly, and thus, the problems that the receiving voice is sharp and the volume is low caused by high signal intensity attenuation of the voice in a low frequency band in the process of using a mobile phone can be solved, and the integral voice effect of the receiving voice can be improved.

In frequency signals obtained by converting input audio signals from time-domain signals to frequency-domain signals, a level control value setting unit 5 establishes a level control value for reducing the levels of spectrums at a noise-components level. A level control value smoothing unit 6 carries out a smoothing process of smoothing the level control value established by the level control value setting unit 5 temporally. A spectral adjustment unit 8 multiplies the level control value after the smoothing process by the frequency signals, performing a level control.

The invention discloses a camera with a built-in speech microphone. The camera comprises a base, a hose connection bar, the built-in speech microphone, a camera main body, a rotary joint, a snapshot button and an anti-jamming device, wherein the base is connected with one end of the hose connection bar, and the other end of the hose connection bar is connected with the camera main body through the rotary joint; the anti-jamming device is arranged in the camera main body; the built-in speech microphone is arranged at the lower end of the camera main body; and the snapshot button is arranged at the upper end of the camera main body. The camera main body can rotate by 360 degrees, the anti-jamming function is strong, a favorable speech sound effect is achieved, and one-click shooting can be realized.

The invention discloses a method and device for realizing stereo call on an ESCO link, a medium and a server, and belongs to the technical field of voice processing. The method for realizing the stereo call on the ESCO link comprises the following steps: establishing Bluetooth communication between a microphone and intelligent terminal equipment through the ESCO link; receiving human voice data and accompaniment data from the intelligent terminal equipment through the microphone; and carrying out sound mixing processing on the human voice data and the accompaniment data through the microphone, and sending a sound mixing processing result to another intelligent terminal equipment. According to the application, the call effect and quality are improved, and the user experience is improved.

The invention provides a voice call-out method, system and device and a storage medium, and the method comprises the steps that a call module carries out the voice call-out of a customer, receives a voice reply of the customer after successful call-out, calls a voice processing engine to generate a machine reply according to the voice reply, and transmits the machine reply to the customer. According to the invention, interaction between a voice outbound machine and a client is realized, interaction behaviors are monitored, outbound standardization and universality which are not based on a specific scene are improved, repeated development is avoided, and machine resources are fully utilized.

The invention relates to the technical field of hearing-aid speech enhancement, and particularly relates to a speech enhancement method and device based on dual-channel neural network time-frequency masking and hearing-aid equipment. The method comprises the steps: receiving a speech signal through employing two microphones under noise and reverberation conditions; performing preliminary speech enhancement on each path of received microphone signals through the trained single-channel neural network, and removing the noise in the same direction as the target speech; different from traditional positioning which blindly depends on signal energy, enabling the self-adaptive ratio mask to provide a view of a target signal, accurately identifying a speech dominant time frequency unit on each microphone channel, using the time frequency units for sound source positioning, and ensuring that high positioning precision is obtained under noise and reverberation conditions; and inputting the calculated weight into a WPD beam former to remove noisy speech in different directions from the target speech and suppress room reverberation so as to obtain enhanced speech with good speech quality and high intelligibility.

The invention relates to a method and a system for dynamically packing speech frames; the method comprises the following steps: both sides transmitting speech frames packet the speech frames in terms of single frame when no blockage occurs in the transmission network, and dynamically adjust the number of packing frames with set step-length, and packet the speech frames with the adjusted number when the blockage occurs in the transmission network; and the system comprises a base station controller and a mobile exchange center. The invention adopts single-frame packing when no blockage occurs in the network, thus guaranteeing no time delay of speech; and the invention adopts multi-frame packing when the blockage occurs, and then adopts single-frame packing again after the blockage is relieved, thus overcoming the inherent defects of single single-frame packing or multi-frame packing, and being capable of maximally optimizing the speech effect on the transmission network.

The invention provides a model training method and device, a voice separation method and device and electronic equipment, and the method comprises the steps: the voice features of a sound signal are input into N pre-trained first neural network models, and N output results are obtained, wherein the N output results are voice features of the voice of the speaker corresponding to N pickup areas separated from the sound signal, and N is an integer greater than 1; voice features of the sound signals are input into a second neural network model, the second neural network model is trained, and a loss function used for training the second neural network model is determined based on the N output results. According to the embodiment of the invention, voice separation is carried out by adopting the trained second neural network model, so that the accuracy of voice separation can be improved.