system
JP2026096422APending Publication Date: 2026-06-15SOFTBANK GROUP CORP
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- SOFTBANK GROUP CORP
- Filing Date
- 2024-12-03
- Publication Date
- 2026-06-15
Smart Images
Figure 2026096422000001_ABST
Abstract
We provide the system. [Solution] A method for constructing a machine learning model that estimates the optimal intonation for a sentence using intonation patterns extracted from past audio data, The machine learning model is provided to the user terminal, and means for generating speech with an intonation suitable for the user's input, A means of outputting the generated audio to the user, A system that includes this.
Need to check novelty before this filing date? Find Prior Art
Claims
[Claim 1] A method for constructing a machine learning model that estimates the optimal intonation for a sentence using intonation patterns extracted from past audio data, The means of providing the aforementioned machine learning model to a user terminal and generating speech with an intonation suitable for the user's input, A means of outputting the generated audio to the user, A system that includes this. [Claim 2] The system according to claim 1, further comprising means for retraining the machine learning model and periodically updating the model. [Claim 3] The system according to claim 1, wherein the user terminal further includes means for generating speech using a speech synthesis engine.