Speed identification conversation device
A speech recognition and speech technology, applied in speech recognition, speech analysis, speech synthesis, etc., can solve the problem of not being able to respond and answer changes in the state and environment, and achieve the effect of more substantial content.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
no. 1 Embodiment
[0049] In this first embodiment, the weighting coefficient is set corresponding to the value of the recognition data of the pre-registered recognition target voice and the value of the change data (time, temperature, weather, date, etc.) that affects the content of the conversation. When the greeting sentence is input It is desired to improve the recognition rate of the greeting sentence. Fig. 1 is a structural diagram illustrating a first embodiment of the present invention. Its structure is briefly explained first, and then each detailed function is explained. In this first embodiment, time is used as an example of change data that affects the content of the conversation.
[0050] In Fig. 1, the interior of the cloth toy dog 30 is provided with a microphone 1, a speech analysis unit 2, a clock unit 3, a coefficient setting unit 4, a speech recognition unit 5, a speech synthesis unit 6, a drive control unit 7, a loudspeaker 8 and a power supply unit 9 etc. The microphone...
no. 2 Embodiment
[0075] Refer below figure 2 A second embodiment of the present invention will be described. figure 2 The cloth toy dog 30 and the operating mechanism 10 etc. which make the mouth of the cloth toy move are omitted. figure 2 The difference between the structure of FIG. 1 and that of FIG. 1 is that a memory 21 is provided for storing the weighting coefficients of each recognizable sentence corresponding to the time data set by the coefficient setting part 4. Other components are the same as in FIG. 1, so the same Parts are marked with the same symbol. In addition, processing between the memory 21 and the coefficient setting unit 4 will be described later.
[0076] exist figure 2 Among them, the voice input from the microphone is analyzed by the voice analysis unit 2 to generate a voice data pattern corresponding to the feature value of the input voice. This speech data pattern is input to the input part of the neural network provided in advance in the speech recognition par...
no. 3 Embodiment
[0090] Refer below image 3 A third embodiment of the present invention will be described. image 3 The cloth toy dog 30 and the operating mechanism 10 etc. which make the mouth of the cloth toy move are omitted. The configuration of the third embodiment includes a microphone 1, a speech analysis unit 2, a clock unit 3, a speech recognition unit 5, a speech synthesis unit 6, a drive control unit 7, a speaker 8, a power supply unit 9, and the like. The microphone 1 inputs voice from the outside, the voice analysis unit 2 analyzes the voice input from the microphone 1, and produces a voice data pattern corresponding to the voice feature value, the clock unit 3 outputs time data, and the voice recognition unit 5 outputs the time data according to the voice output from the above-mentioned voice analysis unit 2. Voice data mode output to the recognition data of above-mentioned input voice, speech synthesis part 6 is based on the recognition data output from this speech recogniti...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More