Evaluation apparatus, evaluation method, and program
View Patent ↗The number of conversational tests required for evaluation of acoustic quality of the ICC system is reduced. An evaluation value converting device 3 evaluates the quality of a conversation made across a near-end acoustic area 100 and a far-end acoustic area 200 inside a vehicle in which a plurality of acoustic areas are predetermined. A voice signal collected by a microphone M 2 disposed in the far-end acoustic area 200 is emitted from a speaker S 1 disposed in the near-end acoustic area 100 . An objective evaluation value acquisition unit 33 acquires an evaluation value using, as an evaluation target sound, a voice signal obtained by adding a voice signal obtained from a first voice signal emitted from a sound source in the far-end acoustic area 200 , collected by the microphone M 2 , and emitted from the speaker S 1 to a voice signal obtained from the first voice signal transmitted through a space inside the vehicle and reaching the near-end acoustic area 100 , with the first voice signal being used as a reference sound.
1. An evaluation apparatus for evaluating quality of a conversation made between a first sound source in a first acoustic area of a plurality of acoustic areas in the vehicle and a second source in a second acoustic area of the plurality of acoustic areas in the vehicle, the evaluation apparatus comprising a processor configured to execute operations comprising:
receiving, by a first microphone in the first acoustic area, a first voice signal originated from the first sound source in the first acoustic area for the conversation, wherein the first acoustic area comprises a first microphone and a first speaker, and the second acoustic area comprises a second microphone and a second speaker, the first acoustic area and the second acoustic area are distinct, and the first voice signal represents a reference sound signal;
transmitting the first voice signal over a communication line between the first microphone and the second speaker;
emitting, by the second speaker in the second acoustic area, a second voice signal, wherein the second voice signal is based on the transmitted first voice signal from the first microphone;
receiving, by the second microphone in the second acoustic area, the second voice signal;
receiving, by the second microphone in the second acoustic area, the first voice signal, wherein the first voice signal has traveled over air from a first speaker in the first acoustic area to the second acoustic area;
generating an evaluation target sound signal by adding the received first voice signal and the received second voice signal; and
acquiring, based at least on the evaluation target sound signal and the reference sound signal, an evaluation value, wherein the evaluation value describes a quality of the conversation across the first acoustic area and the second acoustic area in the vehicle, the evaluation value includes a process to correct a time shift between the evaluation target sound signal and the reference sound signal.
2. The evaluation apparatus according to claim 1 ,
wherein the first speaker is configured to emit a third voice signal based on the second voice signal received by the second microphone, and
the acquiring the evaluation value further comprises using, as the evaluation target sound signal, a fourth voice signal obtained by further adding a fifth voice signal obtained from a sixth voice signal emitted from a second sound source in the second acoustic area, collected by the second microphone in the second acoustic area, and emitted from the first speaker in the first acoustic area.
3. The evaluation apparatus according to claim 2 , the processor further configured to execute operations comprising:
acquiring a simulated voice signal obtained by reflecting, in an aggregated voice signal emitted from the first speaker, a characteristic of transmission between the first speaker and the second acoustic area,
wherein the acquiring further comprises using the simulated voice signal as the aggregated voice signal obtained from the second voice signal emitted from the second sound source in the second acoustic area, collected by the second microphone, and emitted from the first speaker.
4. The evaluation apparatus according to claim 1 , the processor further configured to execute operations comprising:
acquiring a simulated voice signal obtained by reflecting, in a third voice signal obtained from the first voice signal collected by the first microphone, a characteristic of transmission between the first sound source of the first acoustic area and the second acoustic area,
wherein the acquiring further comprises using the simulated voice signal as a fourth voice signal obtained from the first voice signal transmitted through the air inside the vehicle and reaching the second acoustic area.
5. The evaluation apparatus according to claim 1 ,
wherein the evaluation value represents an objective evaluation value relating to the quality of the conversation, and
the processor further configured to execute operations comprising converting the objective evaluation value to a subjective evaluation value.
6. The evaluation apparatus according to claim 5 ,
wherein the converting further comprises linearly converting the objective evaluation value to obtain the subjective evaluation value based on a linear relationship obtained from a subjective evaluation value acquired in a conversational test performed across two acoustic areas of the plurality of acoustic areas and an objective evaluation value acquired using an evaluation target sound and a reference sound obtained by recording a voice of the conversational test.
7. The evaluation apparatus according to claim 5 ,
wherein the converting further comprises obtaining, as the subjective evaluation value, y=ax+b, where x is the objective evaluation value, a is 1.3 or close to 1.3, and b is −0.45 or close to −0.45.
8. An evaluation method for evaluating quality of a conversation made between a first sound source in a first acoustic area and a second sound source in a second acoustic area across the first acoustic area and the second acoustic area in a vehicle, comprising:
receiving, by a first microphone in the first acoustic area, a first voice signal originated from the first sound source in the first acoustic area for the conversation, wherein the first acoustic area comprises a first microphone and a first speaker, and the second acoustic area comprises a second microphone and a second speaker, the first acoustic area and the second acoustic area are distinct, and the first voice signal represents a reference sound signal;
transmitting the first voice signal over a communication line between the first microphone and the second speaker;
emitting, by the second speaker in the second acoustic area, a second voice signal, wherein the second voice signal is based on the transmitted first voice signal from the first microphone;
receiving, by the second microphone in the second acoustic area, the second voice signal;
receiving, by the second microphone in the second acoustic area, the first voice signal, wherein the first voice signal has traveled over air from a first speaker in the first acoustic area to the second acoustic area;
generating, by an evaluation value converting device, an evaluation target sound signal by adding the received first voice signal and the received second voice signal; and
acquiring, by an objective evaluation value acquisition unit, based at least on the evaluation target sound signal and the reference sound signal, an evaluation value, wherein the evaluation value describes a quality of the conversation across the first acoustic area and the second acoustic area in the vehicle, the evaluation value includes a process to correct a time shift between the evaluation target sound signal and the reference sound signal.
9. The evaluation method according to claim 8 ,
wherein the first speaker is configured to emit a third voice signal based on the second voice signal received by the second microphone, and
the acquiring further comprises using, as the evaluation target sound signal, a fourth voice signal obtained by further adding a fifth voice signal obtained from a sixth second voice signal emitted from a second sound source in the second acoustic area, collected by the second microphone in the second acoustic area, and emitted from the first speaker in the first acoustic area.
10. The evaluation method according to claim 9 , further comprising:
acquiring a simulated voice signal obtained by reflecting, in an aggregated voice signal emitted from the first speaker, a characteristic of transmission between the first speaker and the second acoustic area,
wherein the acquiring the evaluation value further comprises using the simulated voice signal as the aggregate voice signal obtained from the second voice signal emitted from the second sound source in the second acoustic area, collected by the second microphone, and emitted from the first speaker.
11. The evaluation method according to claim 8 , further comprising:
acquiring a simulated voice signal obtained by reflecting, in a third voice signal obtained from the first voice signal collected by the first microphone, a characteristic of transmission from the first sound source of the first acoustic area to the second acoustic area,
wherein the acquiring the evaluation value further comprises using the simulated voice signal as a fourth voice signal obtained from the first voice signal transmitted through the air inside the vehicle and reaching the second acoustic area.
12. The evaluation method according to claim 8 ,
wherein the evaluation value represents an objective evaluation value relating to the quality of the conversation, and
the evaluation method further comprising:
converting the objective evaluation value to a subjective evaluation value.
13. The evaluation method according to claim 12 ,
wherein the converting further comprises linearly converting the objective evaluation value to obtain the subjective evaluation value based on a linear relationship obtained from a subjective evaluation value acquired in a conversational test performed across two acoustic areas of the plurality of acoustic areas and an objective evaluation value acquired using an evaluation target sound and a reference sound obtained by recording a voice of the conversational test.
14. The evaluation method according to claim 12 ,
wherein the converting further comprises obtaining, as the subjective evaluation value, y=ax+b, where x is the objective evaluation value, a is 1.3 or close to 1.3, and b is −0.45 or close to −0.45.
15. A non-transitory computer-readable recording medium storing computer-executable evaluation program instructions for performing the evaluation method according to claim 8 .