Emotion recognition in speech chatbot job interview system

1. A method, comprising:

selecting, by a computer system with one or more processors coupled with at least one memory unit, a first question from a question bank based on one or more selection criteria;

obtaining, by the computer system, a sentiment classifier of an audio clip of a first answer to the first question, wherein the sentiment classifier is derived from an emotion classifier resulting from a convolutional neural network (CNN) model analysis of the audio clip;

obtaining, by the computer system, an assessment result to the first question by analyzing the audio clip of the first answer to the first question using a recurrent neural network (RNN) model; and

generating a first emotion response result to the first question based on the sentiment classifier and the assessment result, wherein the first emotion response result presents a sampling experience factor to the assessment result.

2. The method of claim 1 , wherein the emotion classifier is one selecting from an emotion group comprising angry emotion, excited emotion, frustrated emotion, happy emotion, neutral emotion, sad emotion, and surprised emotion, and the sentiment classifier is one selecting from a sentiment group comprising extremely positive, positive, neutral, negative, extremely negative, and surprised.

3. The method of claim 2 , wherein the sentiment classifier is mapped to the emotional classifier.

4. The method of claim 1 , wherein the first question is a personal trait question, wherein the first answer to the personal trait question identifies one or more personal traits.

5. The method of claim 4 , wherein the assessment result of the personal trait question reveals a positive or a negative propensity on a scaling basis to the one or more personal traits identified by the first question.

6. The method of claim 5 , wherein the sentiment classifier alters the assessment result when the sentiment classifier indicates a positive, an extremely positive, a negative or an extremely negative result.

7. The method of claim 1 , wherein the first question is a technical question, wherein the first answer to technical question identifies one or more technical skills on a scaling basis.

8. The method of claim 7 , wherein the sentiment classifier serves as a reference factor to the assessment result, wherein the assessment result for the first question is correct when the sentiment classifier indicates an extremely negative, a negative or a neutral result, the first emotion response result for the first question indicates the first question is a below-skill-level question.

9. The method of claim 1 , further comprising: selecting a second question from the question bank, wherein a selection criterion is based on the first emotion response result.

10. The method of claim 1 , further comprising: presenting the first question with a speech chatbot.

11. The method of claim 10 , wherein a voice of the speech chatbot is adjustable based on one or more dynamically configured adjustment factors.

12. The method of claim 11 , further comprising selecting a second question from the question bank, wherein one adjustment factor is the first emotion response result.

13. An apparatus comprising:

a network interface that connects the apparatus to a communication network;

a memory; and

one or more processors coupled to one or more memory units, the one or more processors configured to

select a first question from a question bank based on one or more selection criteria;

obtain a sentiment classifier of an audio clip of a first answer to the first question, wherein the sentiment classifier is derived from an emotion classifier resulting from a convolutional neural network (CNN) model analysis of the audio clip;

obtain an assessment result to the first question by analyzing the audio clip of the first answer to the first question using a recurrent neural network (RNN) model; and

generate a first emotion response result to the first question based on the sentiment classifier and the assessment result, wherein the first emotion response result presents a sampling experience factor to the response assessment result.

14. The apparatus of claim 13 , wherein the emotion classifier is one selecting from an emotion group comprising angry emotion, excited emotion, frustrated emotion, happy emotion, neutral emotion, sad emotion, and surprised emotion, and the sentiment classifier is mapped to corresponding emotion classifier and is one selecting from a sentiment group comprising extremely positive, positive, neutral, negative, extremely negative, and surprised.

15. The apparatus of claim 13 , wherein the first question is a personal trait question, wherein the first answer to the personal trait question identifies one or more personal traits and reveals a positive or a negative propensity on a scaling basis to the one or more personal traits identified by the first question.

16. The apparatus of claim 15 , wherein the sentiment classifier alters the assessment result when the sentiment classifier indicates a positive, an extremely positive, a negative or an extremely negative result.

17. The apparatus of claim 13 , wherein the first question is a technical question, wherein the first answer to technical question identifies one or more technical skills on a scaling basis.

18. The apparatus of claim 13 , wherein the one or more processors are further configured to select a second question from the question bank, wherein a selection criterion is based on the first emotion response result.

19. The apparatus of claim 13 , wherein the one or more processors are further configured to present the first question with a speech chatbot, and wherein a voice of the speech chatbot is adjustable based on one or more dynamically configured adjustment factors.

20. The apparatus of claim 19 , wherein the one or more processors are further configured to select a second question from the question bank, wherein one adjustment factor is the first emotion response result.