Voice outputting method and device, voice shielding method and device

[0048] This embodiment provides a voice output method, such as figure 1 As shown, the steps are as follows:

[0049] Step 100, input a voice as an input voice;

[0050] Step 101, preprocessing the input speech and obtaining waveform characteristic data of the input speech;

[0051] Step 102, judging whether the waveform feature data matches the waveform feature data of a voice in a reserved voice bank, if so, then enter step 103, if not, enter step 104;

[0052] Step 103, calling the extended voice corresponding to the waveform feature data of the voice in the reserved voice library;

[0053] Step 104, outputting the input voice;

[0054] Step 105, synchronously output the input speech and the extended speech.

[0055] Obtain its waveform feature data after preprocessing the input voice, and then match the waveform feature data with the waveform feature data of a voice in the reserved voice bank, and when the matching is successful, the input voice and the reserved voice are sy

[0063] This embodiment provides a voice shielding method, such as image 3 As shown, the steps are as follows:

[0064] Step 100, input a voice as an input voice;

[0065] Step 101, preprocessing the input speech and obtaining waveform characteristic data of the input speech;

[0066] Step 102, judging whether the waveform feature data matches the waveform feature data of a voice in a reserved voice bank, if so, then enter step 103, if not, enter step 104;

[0067] Step 103, calling the extended voice corresponding to the waveform feature data of the voice in the reserved voice database;

[0068] Step 104, outputting the input voice;

[0069] Step 105, while outputting the extended voice, shielding the output of the input voice.

[0070] The difference between this embodiment and the voice output method in Embodiment 1 is that in step 105, the input voice and the extended voice are synchronously output in Embodiment 1, and in this embodiment, by closing the channel for transm