Information processing device, information processing method, program, and recording medium

Abstract

Provided is an information processing device which controls sound volume so that a user can reliably discern speech such as speech guidance without incompatibility in terms of the sense of hearing.　The information processing device comprises: a noise detection unit 20 which detects the noise level of a sound in a vehicle cabin, the sound being picked up in a sound pickup unit 10; a command speaking detection unit 30 which detects whether a command speaking period is in effect, which is a period after a wakeup word from spoken speech of the user is detected to when command speaking from the user is detected; and a sound volume control unit 50 which, when the noise level is not detected in the command speaking period, sets, on the basis of the noise level, the volume of a speech output to the user, and, when the noise level is detected in the command speaking period, does not set the volume of speech based on the noise level.

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing method, a program, and a recording medium. [Background technology]

[0002] In general, a navigation device provides voice guidance, such as route information for the next intersection where the user should turn and traffic congestion information for the road on which the user plans to travel, to guide the user along a route to a destination set by the user. With this type of device, voice guidance is provided at a volume selected by the user from several predetermined volume levels, but road noise and engine noise while driving can increase the noise level, making it difficult for the user to hear the voice guidance or to misinterpret the guidance. A technique has been disclosed for controlling the volume of audio guidance based on noise levels inside and outside the vehicle so that the audio guidance can be heard even in noisy driving environments (see, for example, Patent Document 1). The above-mentioned prior art proposes using a microphone placed inside the vehicle to detect the level of noise generated inside and outside the vehicle, and controlling the volume of the voice guidance based on the level of that noise. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2002-91488 Summary of the Invention [Problem to be solved by the invention]

[0004] In recent years, devices have become widespread in which a voice assistant function is activated when a user speaks a wake-up word (also called a wake word or hot word), and the device's operation instructions or information search are performed based on the user's speech. In the case of navigation devices equipped with the voice assistant function described above, the voice spoken by the user to give operating instructions, etc. is picked up as noise and the volume of the voice guidance is controlled, which can result in the voice guidance being output at a loud volume or the volume of the voice guidance changing mid-output, making the voice guidance difficult to hear, for example.

[0005] The present invention has been made in consideration of the problems cited as examples above, and its main object is to provide an information processing device, an information processing method, a program, and a recording medium that control the volume so that a user can reliably hear voice guidance and other sounds without feeling uncomfortable when listening. [Means for solving the problem]

[0006] In order to solve the above problem, the invention of claim 1 includes a noise detection unit that detects a noise level, a command utterance detection unit that detects whether or not a command utterance period is occurring, which is a period from when a wake-up word is detected in a user's speech to when a command utterance from the user is detected, and a noise detection unit that detects whether or not the noise level is occurring. Detection of is the command utterance period in If not, the volume of the sound to be output to the user is set based on the noise level, and the noise level Detection of is the command utterance period is In this case, the information processing device is equipped with a volume control unit that sets the volume of the audio based on the noise level obtained before a predetermined time before the time when the wake-up word was detected.

[0007] The invention of claim 6 is an information processing method for an information processing device including a noise detection unit, a command utterance detection unit, and a volume control unit, the information processing method including a first step in which the noise detection unit detects a noise level, a second step in which the command utterance detection unit detects whether or not it is a command utterance period, which is a period from detecting a wake-up word from a user's speech to detecting a command utterance from the user, and a volume control unit in which the volume control unit detects the noise level. Detection ofis the command utterance period in If not, the volume of the sound to be output to the user is set based on the noise level, and the noise level Detection of is the command utterance period is In this case, the information processing method includes a third step of setting the volume of the audio based on the noise level obtained before a predetermined time before the time when the wake-up word was detected.

[0008] The invention described in claim 7 is a program for causing a computer to execute an information processing method for an information processing device including a noise detection unit, a command utterance detection unit, and a volume control unit, the program including a first step in which the noise detection unit detects a noise level; a second step in which the command utterance detection unit detects whether or not a command utterance period is occurring, which is a period from when a wake-up word is detected from a user's speech to when a command utterance is detected from the user; and a volume control unit in which the volume control unit detects the noise level. Detection of is the command utterance period in If not, the volume of the sound to be output to the user is set based on the noise level, and the noise level Detection of is the command utterance period is and a third step of setting the volume of the audio based on the noise level obtained a predetermined time before the time the wake-up word was detected.

[0009] The invention described in claim 8 is a non-transitory computer-readable recording medium having recorded thereon a program for causing a computer to execute an information processing method for an information processing device including a noise detection unit, a command utterance detection unit, and a volume control unit, the information processing method including a first step in which the noise detection unit detects a noise level, and a second step in which the command utterance detection unit detects a wake-up word from a user's speech until detecting a command utterance from the user. It is a period a second step of detecting whether or not it is a command utterance period, and Detection ofis the command utterance period in If not, the volume of the sound to be output to the user is set based on the noise level, and the noise level Detection of is the command utterance period is and a third step of setting the volume of the audio based on the noise level obtained a predetermined time before the time the wake-up word was detected. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a diagram illustrating a configuration of an information processing device according to a first embodiment. [Figure 2] FIG. 4 is a timing chart of the volume control unit of the information processing device according to the first embodiment. [Figure 3] FIG. 4 is a diagram illustrating a processing flow of a volume control unit of the information processing device according to the first embodiment. [Figure 4] 4 is a diagram illustrating a table that is referred to when a volume control unit of the information processing device according to the first embodiment determines a volume. FIG. [Figure 5] FIG. 10 is a diagram illustrating a configuration of an information processing device according to a second embodiment. [Figure 6] FIG. 10 is a timing chart of a volume control unit of the information processing device according to the second embodiment. [Figure 7] FIG. 10 is a diagram illustrating a processing flow of a volume control unit of the information processing device according to the second embodiment. [Figure 8] FIG. 10 is a diagram illustrating a configuration of an information processing device according to a third embodiment. [Figure 9] FIG. 11 is a timing chart of a volume control unit of the information processing device according to the third embodiment. [Figure 10] FIG. 11 is a diagram illustrating a processing flow of a volume control unit of the information processing device according to the third embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0011] The information processing device of this embodiment includes a noise detection unit that detects the noise level of sounds within the vehicle cabin, a command utterance detection unit that detects whether or not the current period is a command utterance period, which is the period from when a wake-up word is detected from the user's speech to when a command utterance from the user is detected, and a volume control unit that sets the volume of the voice to be output to the user based on the noise level if the noise level is not detected during the command utterance period, and does not set the volume of the voice based on the noise level if the noise level is detected during the command utterance period. The noise detection unit detects the volume of noise (noise level) inside the vehicle. The noise detection unit also transmits the detected noise level and information as to whether or not the noise level was detected during a command utterance period, which will be described later, to the volume control unit. Specifically, when detecting a noise level, the noise detection unit checks the information received from the command utterance detection unit, determines whether a noise level was detected during the command utterance period, and transmits the noise level and the determination result to the volume control unit. The command utterance detection unit detects whether or not the current time is a command utterance period, which is the period from when a wake-up word is detected in a user's voice to when a command utterance is detected. Specifically, the command utterance detection unit detects the wake-up word spoken by the user from the sounds picked up inside the vehicle cabin, and the command utterances that follow the wake-up word, such as instructions to operate the device or instructions to search for information, and detects whether the current period is one in which the command is being spoken. The command utterance detection unit transmits the detection result of whether or not it is a command utterance period to the noise detection unit. If the noise level detected by the noise detection unit is not detected during the command utterance period, the volume control unit sets the volume of the voice to be output to the user based on the detected noise level. Here, the audio output to the user includes at least one of an audio guidance sound for guiding the user along a route to a destination set by the user and an audio guidance sound for recommendation information for the user. On the other hand, if the noise level detected by the noise detection unit is detected during the command utterance period, the volume control unit does not set the volume of the voice to be output to the user. The noise level detected during the command utterance period may be high because it includes the sound of the command utterance spoken by the user. Therefore, if the noise level detected by the noise detection unit is detected during the command utterance period, the volume control unit does not set the volume of the voice. This eliminates volume changes during voice output due to user commands and voice output with an unnecessarily high volume, allowing the user to hear without any discomfort. In addition, if the noise level detected by the noise detection unit is not detected during the command utterance period, the volume control unit controls the volume of the voice output to the user based on the detected noise level. That is, the volume control unit sets the volume of the voice based on the noise level of sounds that do not include the user's command utterance. As a result, the sound is output at an appropriate volume that matches the noise level inside the vehicle, allowing the user to reliably hear the sound without any discomfort to the ear.

[0012] Example 1 An information processing device 1 according to this embodiment will be described with reference to FIGS.

[0013] <Configuration of information processing device 1> The configuration of an information processing device 1 according to this embodiment will be described with reference to FIG. The information processing device 1 includes at least a sound collection unit 10, a noise detection unit 20, a command utterance detection unit 30, a voice output unit 40, and a volume control unit 50.

[0014] The sound collection unit 10 is configured by, for example, a microphone, and transmits the collected sound to the noise detection unit 20 and the command utterance detection unit 30. Specifically, the sound collection unit 10 collects sound using a microphone installed inside the vehicle, and transmits the collected sound to the noise detection unit 20 and the command utterance detection unit 30. Examples of sounds picked up by the sound pickup unit 10 include the user's speech, environmental sounds around the vehicle, engine noise, wind noise, road noise, air conditioner operating sounds, and voice guidance output from the speaker. In addition, since the sound collection unit 10 is only required to collect sounds generated inside and outside the vehicle as exemplified, it may be configured using, for example, a microphone for hands-free calls or a microphone for inputting voice instructions that is installed inside the vehicle.

[0015] The noise detection unit 20 detects the noise level of the sound collected by the sound collection unit 10. The noise detection unit 20, for example, divides the sound received from the sound collection unit 10 into predetermined time intervals (for example, every second), calculates the average noise level for each divided sound, and transmits the calculated value to the volume control unit 50, which will be described later. Furthermore, the noise detection unit 20 transmits the detected noise level and information as to whether or not the noise level was detected during the command utterance period to the volume control unit 50, which will be described later. Specifically, when detecting a noise level, the noise detection unit 20 checks the information received from the command utterance detection unit 30 described later, determines whether a noise level has been detected during the command utterance period, and transmits the noise level and the determination result together to the volume control unit 50.

[0016] The command utterance detection unit 30 detects whether or not it is a command utterance period, which is a period from when a wake-up word is detected in the user's voice to when a command utterance from the user is detected. Specifically, the command utterance detection unit 30 detects a wake-up word uttered by the user from the sound received from the sound collection unit 10. Furthermore, the command utterance detection unit 30 detects a command utterance uttered by the user after detecting the wake-up word. That is, the command utterance detection unit 30 determines the period from when the wake-up word is detected until when the command utterance is detected as the command utterance period, and transmits the determination result to the noise detection unit 20. Here, the wake-up word is a word that the user first utters when giving an instruction to operate the information processing device 1, an instruction to execute an information search, or the like, by voice. When the wake-up word is spoken, a voice assistant unit (not shown) included in the information processing device 1 is activated, and the voice assistant unit executes processing in accordance with the content of the command spoken by the user. As described above, the wake-up word is a short phrase that the user speaks to activate the voice assistant unit, such as "Hey Pioneer." Furthermore, a command utterance is a specific instruction or other word that the user utters after uttering the wake-up word described above and activating the voice assistant unit, and examples include "Tell me what the weather will be like tomorrow" or "Tell me the way to my house."

[0017] The audio output unit 40 is configured by, for example, an amplifier and a speaker, and outputs audio to the user. The audio (hereinafter referred to as system audio) output to the user by the audio output unit 40 includes at least one of audio guidance sounds (such as advance route information for the next intersection to turn at) to guide the user along the route to the destination set by the user, and audio guidance sounds of recommended information for the user. Here, examples of system sounds include a startup sound such as "beep" or a warning sound that is output when audio guidance such as route guidance starts. Furthermore, voices such as responses from the voice assistant unit described above can also be exemplified as system sounds. Furthermore, examples of the above-mentioned recommendation information include information on recommended tourist spots, shops, etc. in the vicinity of the vehicle's position or destination. Based on the volume setting received from the volume control unit 50 described later, the audio output unit 40 outputs audio such as audio guidance received from a navigation control unit (not shown) to guide the user along the route to the destination set by the user (such as advance route information for the next intersection to turn at) and audio guidance containing recommendation information for the user (such as information on recommended tourist spots and shops in the vicinity of the vehicle's position or destination). Furthermore, the audio output unit 40 outputs a response voice in response to a command uttered by the user, received from a voice assistant unit (not shown), based on a volume setting received from a volume control unit 50 (described later).

[0018] The volume control unit 50 will be described with reference to FIG. If the noise level received from the noise detection unit 20 is not detected during the command utterance period, the volume control unit 50 sets the volume of the system sound based on the noise level, and if the noise level is detected during the command utterance period, the volume control unit 50 does not set the volume of the audio based on the noise level.

[0019] As shown in FIG. 2, the volume control unit 50 acquires the noise levels (N1, N2, ...) detected by the noise detection unit 20 at predetermined time intervals (for example, every second), and determines the volume settings (S1, S2, ...) of the system sounds based on the noise levels.

[0020] If the noise level received from the noise detection unit 20 is not detected during the command utterance period (e.g., N1, N2, N3, N7), the volume control unit 50 transmits a volume setting value based on the noise level detected by the noise detection unit 20 to the audio output unit 40 (e.g., S1, S2, S3, S7). On the other hand, when the noise level received from the noise detection unit 20 is detected during the command utterance period (N4 to N6), the volume control unit 50 does not set the volume setting (S4 to S6) based on the noise level received from the noise detection unit 20 to the audio output unit 40. That is, the noise levels (N4 to N6) detected during the command utterance period are noise levels that include the voice of the command utterance by the user, and therefore there is a possibility that the noise levels are high. Therefore, the volume control unit 50 does not transmit to the audio output unit 40 the volume settings (S4 to S6) based on the noise levels (N4 to N6) acquired during this period. During the period when the volume settings (S4 to S6) are not transmitted to the audio output unit 40, the system sound is output according to the volume setting (S3) that was last transmitted to the audio output unit 40 by the volume control unit 50. The processing flow of the volume control unit 50 will be described in detail below.

[0021] <Processing of the volume control unit 50> The processing flow of the volume control unit 50 will be described in detail with reference to FIGS.

[0022] As shown in FIG. 3, the volume control unit 50 determines whether or not the ACC power supply (accessory power supply) of the vehicle is in an ON state (step S100). If it is determined that the ACC power supply of the vehicle is not in an ON state ("NO" in step S100), the process returns to step S100 and transitions to a standby state. On the other hand, if it is determined that the ACC power supply of the vehicle is in the ON state ("YES" in step S100), the process proceeds to step S110.

[0023] If it is determined that the ACC power supply of the vehicle is in an on state ("YES" in step S100), the volume setting to be set in the audio output unit 40 is initialized (step S110). Immediately after the information processing device 1 is started up, the noise detection unit 20 has not yet completed detection of the noise level, so the volume control unit 50 transmits the volume setting selected by the user to the audio output unit 40. Here, the volume setting selected by the user is the volume setting selected by the user while listening to the system sound in a low noise state (for example, when the engine is on and the vehicle is stopped). The volume setting set by the user is stored in a memory (not shown).

[0024] The volume control unit 50 acquires the detected noise level and information on whether or not the noise level was detected during the command utterance period from the noise detection unit 20 (step S120), and then proceeds to step S130.

[0025] The volume control unit 50 determines whether the noise level acquired from the noise detection unit 20 was detected during the command utterance period (step S130). If it is determined that the noise level acquired from the noise detection unit 20 has not been detected during the command utterance period ("NO" in step S130), the process proceeds to step S140. On the other hand, if it is determined that the noise level acquired from the noise detection unit 20 was detected during the command utterance period ("YES" in step S130), the process proceeds to step S150. In other words, if it is determined that the noise level obtained from the noise detection unit 20 is detected during the command utterance period, the noise level detected includes the voice of the user's command utterance, and therefore the volume control unit 50 does not set the volume based on the noise level.

[0026] If the volume control unit 50 determines that the noise level acquired from the noise detection unit 20 has not been detected during the command utterance period ("NO" in step S130), it determines the volume setting for the system sound based on the acquired noise level and transmits the volume setting to the audio output unit 40 (step S140).

[0027] Here, a method for the volume control unit 50 to determine the volume setting of the system sound based on the noise level acquired from the noise detection unit 20 will be described. As shown in FIG. 4, the volume control unit 50 determines the volume setting to be set in the audio output unit 40 based on the amount of volume correction associated with the noise level acquired from the noise detection unit 20. Specifically, the volume control unit 50 corrects the volume set in step S110 (the volume of the system sound set in advance by the user) according to the volume correction amount linked to the noise level. For example, if the noise level acquired from the noise detection unit 20 is greater than 60 dB and equal to or less than 65 dB, the volume control unit 50 transmits a volume setting to the audio output unit 40 such that the volume set in step S110 is +6 dB. In FIG. 4, the noise level value obtained from the noise detection unit 20 is divided into four stages to determine the volume correction value, but the volume of the system sound may be controlled by dividing the value into more stages.

[0028] The volume control unit 50 determines whether the ACC power supply (accessory power supply) of the vehicle is in an ON state (step S150). If it is determined that the ACC power supply of the vehicle is in the ON state ("YES" in step S150), the process proceeds to step S120 and continues. On the other hand, if it is determined that the ACC power supply of the vehicle is not in an ON state ("NO" in step S150), the process is ended.

[0029] The information processing device 1 of this embodiment includes a noise detection unit 20 that detects the noise level of sounds within the vehicle cabin, a command utterance detection unit 30 that detects whether or not the command utterance period is occurring, which is the period from when a wake-up word is detected from the user's speech to when a command utterance from the user is detected, and a volume control unit 50 that sets the volume of the sound (system sound) to be output to the user based on the noise level if the noise level is not detected during the command utterance period, and does not set the volume of the sound (system sound) based on the noise level if the noise level is detected during the command utterance period. The noise detection unit 20 detects the volume of noise (noise level) inside the vehicle. Furthermore, the noise detection unit 20 transmits to the volume control unit 50 the detected noise level and information as to whether or not the noise level was detected during the command utterance period. Specifically, when detecting a noise level, the noise detection unit 20 checks the information received from the command utterance detection unit 30, determines whether a noise level has been detected during the command utterance period, and transmits the noise level and the determination result to the volume control unit 50. The command utterance detection unit 30 detects whether or not the current time is a command utterance period, which is the period from when a wake-up word is detected from the user's speech to when a command utterance from the user is detected. Specifically, the command utterance detection unit 30 detects the wake-up word spoken by the user and the command utterances following the wake-up word, such as device operation instructions or information search instructions, from the sounds within the vehicle cabin picked up by the sound collection unit 10, and detects whether the current period is one in which a command is being spoken. The command utterance detection unit 30 transmits the detection result indicating whether or not it is a command utterance period to the noise detection unit 20 . If the noise level detected by the noise detection unit 20 is not detected during the command utterance period, the volume control unit 50 sets the volume of the system sound based on the detected noise level. On the other hand, if the noise level detected by the noise detection section 20 is detected during the command utterance period, the volume control section 50 does not set the volume of the system sound based on the noise level. The noise level detected during the command utterance period may be high because it includes the sound of the command utterance spoken by the user. Therefore, if the noise level detected by the noise detection unit 20 is detected during the command utterance period, the volume control unit 50 does not set the volume of the voice. This eliminates volume changes during voice output due to user command utterances, and voice output with unnecessarily increased volume, so the user can reliably hear what is being said without any sense of discomfort to the ear. Furthermore, if the noise level detected by the noise detection unit 20 is not detected during the command utterance period, the volume control unit 50 controls the volume of the system sound based on the detected noise level. That is, the volume control unit 50 sets the volume of the system sound based on the noise level of sounds that do not include command utterances from the user. This allows the system sound to be output at an appropriate volume that matches the noise level inside the vehicle, allowing the user to reliably hear the system sound without any discomfort to the ear.

[0030] In addition, the volume control unit 50 of the information processing device 1 according to this embodiment performs volume correction control in accordance with the noise level inside the vehicle, so that even in noisy environments, the system sound is output so that it can be heard at a volume close to the volume that the user has previously set. This allows the user to reliably hear the system sound even in noisy environments.

[0031] Furthermore, the volume control unit 50 of the information processing device 1 according to this embodiment transmits the volume setting based on the noise level to the audio output unit 40 even when the system sound is not being output. As a result, even if the volume control unit 50 is configured to be unable to grasp the output timing of system sounds such as voice guidance, the volume setting based on the noise level is always transmitted to the audio output unit 40, so that the user can reliably hear the system sound without any audible discomfort no matter when the system sound is output.

[0032] Furthermore, the volume control unit 50 of the information processing device 1 according to this embodiment determines the volume setting of the system sound at the timing when the noise level is received from the noise detection unit 20, and transmits the volume setting to the audio output unit 40. When transmitting the volume setting to the audio output unit 40 in accordance with the timing at which the system sound is output, it is necessary to temporarily store the volume setting value determined based on the noise level in memory. However, since the volume control unit 50 does not need to set the volume in accordance with the timing at which the system sound is output, it is possible to eliminate the memory for storing the volume setting.

[0033] <Example 2> An information processing device 1A according to this embodiment will be described with reference to FIGS. Note that components with the same reference numerals as those in the first embodiment have the same functions, and therefore detailed descriptions thereof will be omitted.

[0034] <Configuration of information processing device 1A> The configuration of an information processing device 1A according to this embodiment will be described with reference to FIG. The information processing device 1A includes at least a sound collection unit 10, a noise detection unit 20, a command utterance detection unit 30A, a voice output unit 40, and a volume control unit 50A.

[0035] The command utterance detection unit 30A detects whether or not it is a command utterance period, which is a period from when a wake-up word is detected from the user's voice to when a command utterance from the user is detected. Specifically, the command utterance detection unit 30A detects, from the sound received from the sound collection unit 10, a wake-up word uttered by the user. Furthermore, the command utterance detection unit 30A detects a command utterance uttered by the user after detecting the wake-up word. That is, the command utterance detection unit 30A determines the period from when the wake-up word is detected until when the command utterance is detected as the command utterance period, and transmits the determination result to the noise detection unit 20.

[0036] Furthermore, when the command utterance detection unit 30A detects a wake-up word, it transmits the utterance time required for the user to utter the wake-up word to the volume control unit 50A. Specifically, when the command utterance detection unit 30A detects a wake-up word spoken by the user from the sounds inside the vehicle received from the sound collection unit 10, it detects the time it took for the user to speak the wake-up word and transmits the time information (utterance time) to the volume control unit 50A. More specifically, the time (e.g., 850 msec) taken for the user to speak the wake-up word spoken by the user, for example, the words "Hey Pioneer," is detected, and the time information (speech time) is sent to the volume control unit 50A.

[0037] The volume control unit 50A will be described with reference to FIG. If the noise level received from the noise detection unit 20 is not detected during the command utterance period, the volume control unit 50A sets the volume of the system sound based on the noise level, and if the noise level is detected during the command utterance period, the volume control unit 50A does not set the volume of the audio based on the noise level. Furthermore, when the volume control unit 50A receives the utterance time from the command utterance detection unit 30A, it sets the volume of the system sound based on the noise level acquired at least up to the time point prior to that time point.

[0038] As shown in FIG. 6, the volume control unit 50A receives noise levels (N1, N2, ...) detected by the noise detection unit 20 at predetermined time intervals (e.g., every second), and determines the volume setting (S1, S2, ...) of the system voice based on the noise levels.

[0039] If the noise level received from the noise detection unit 20 is not detected during the command utterance period (e.g., N1, N2, N3, N7), the volume control unit 50A transmits a volume setting value based on the noise level detected by the noise detection unit 20 to the audio output unit 40 (e.g., S1, S2, S3, S7). On the other hand, when the noise level received from the noise detection unit 20 is detected during the command utterance period (N4 to N6), the volume control unit 50A does not transmit the volume setting (S4 to S6) based on the noise level received from the noise detection unit 20 to the audio output unit 40. That is, the noise levels (N4 to N6) detected during the command utterance period may be high because they include the voice of the command utterance by the user. Therefore, the volume control unit 50A does not set the volume settings (S4 to S6) in the audio output unit 40 based on the noise levels (N4 to N6) acquired during this period.

[0040] As shown in Figures 6(c) to (e), when the volume control unit 50A receives the speech time, which is information about the time required to utter the wake-up word, from the command utterance detection unit 30A, it transmits to the audio output unit 40 the volume setting (S2) for the system sound based on the noise level (N2) obtained before the point in time counted back in time from that point in time. The sound picked up between the time when the speech time information is received from the command speech detection unit 30A and the time going back in time from the time when the speech time information is received includes the sound of the wake-up word spoken by the user, which may increase the noise level. Therefore, the volume setting of the system sound based on the noise level before the wake-up word is spoken is sent to the audio output unit 40. Specifically, the noise level (N3) detected by the noise detection unit 20 is not the noise level detected during the command utterance period, but is a noise level detected including the sound of the wake-up word spoken by the user, so the volume control unit 50A transmits to the audio output unit 40 a volume setting (S2) based on the noise level (N2) obtained before the point in time going back in time during which the command was uttered. More specifically, the volume control unit 50A reads from a memory (not shown) the noise level acquired before the point in time prior to the time when the speech time information was received from the command speech detection unit 30A, and transmits the volume setting (S2) of the system sound based on that noise level (N2) to the audio output unit 40. The processing flow of the volume control unit 50A will be described in detail below.

[0041] <Processing of the volume control unit 50A> The processing flow of the volume control unit 50A will be described in detail with reference to FIG.

[0042] As shown in FIG. 7, the volume control unit 50A determines whether the ACC power supply (accessory power supply) of the vehicle is in an ON state (step S100). If it is determined that the ACC power supply of the vehicle is not in an ON state ("NO" in step S100), the process returns to step S100 and transitions to a standby state. On the other hand, if it is determined that the ACC power supply of the vehicle is in the ON state ("YES" in step S100), the process proceeds to step S110.

[0043] If it is determined that the ACC power supply of the vehicle is in an on state ("YES" in step S100), the volume setting to be set in the audio output unit 40 is initialized (step S110).

[0044] The volume control unit 50A acquires the detected noise level and information on whether the noise level was detected during the command utterance period from the noise detection unit 20, and stores the information in a memory (not shown) (step S200). Here, the volume control unit 50A stores the detection information acquired from the noise detection unit 20 in a memory (not shown) in order to keep a history of the detection information received from the noise detection unit 20 in chronological order. When the detected information is stored, time information or the like is added to each piece of information and stored so that it is possible to determine when the information was received from the noise detection unit 20.

[0045] The volume control unit 50A determines whether or not information on the utterance time of the wake-up word has been received from the command utterance detection unit 30A (step S210). If it is determined that information on the utterance time of the wake-up word has been received from the command utterance detection unit 30A ("YES" in step S210), the process proceeds to step S220. On the other hand, if it is determined that information on the utterance time of the wake-up word has not been received from the command utterance detection unit 30A ("NO" in step S200), the process proceeds to step S130.

[0046] If the volume control unit 50A determines that it has received information on the utterance time of the wake-up word from the command utterance detection unit 30A ("YES" in step S210), it sets the volume of the system sound based on the noise level detected before the time point at which the utterance time was received (step S220), and transitions to step S130. Specifically, when the volume control unit 50A receives an utterance time from the command utterance detection unit 30A, it reads from a memory (not shown) the noise level last obtained before the point in time counted back in utterance time from that point in time, and transmits the volume setting for the system sound based on that noise level to the audio output unit 40.

[0047] The volume control unit 50A determines whether the noise level acquired from the noise detection unit 20 was detected during the command utterance period (step S130). If it is determined that the noise level acquired from the noise detection unit 20 has not been detected during the command utterance period ("NO" in step S130), the process proceeds to step S140. On the other hand, if it is determined that the noise level acquired from the noise detection unit 20 was detected during the command utterance period ("YES" in step S130), the process proceeds to step S150. In other words, if it is determined that the noise level obtained from the noise detection unit 20 is detected during the command utterance period, the noise level detected includes the voice of the user's command utterance, and therefore the volume control unit 50A does not set the volume based on the noise level.

[0048] If it is determined that the noise level acquired from the noise detection unit 20 was not detected during the command utterance period ("NO" in step S130), the volume setting for the system sound is determined based on the acquired noise level, and the volume setting is sent to the audio output unit 40 (step S140). The method for determining the volume setting of the system sound based on the noise level acquired from the noise detection unit 20 is the same as in the first embodiment.

[0049] The volume control unit 50A determines whether the ACC power supply (accessory power supply) of the vehicle is in the ON state (step S150). If it is determined that the ACC power supply of the vehicle is in the ON state ("YES" in step S150), the process proceeds to step S120 and continues. On the other hand, if it is determined that the ACC power supply of the vehicle is not in an ON state ("NO" in step S150), the process is ended.

[0050] In the information processing device 1A of this embodiment, when the command utterance detection unit 30A detects a wake-up word, it transmits the speech time required for the user to utter the wake-up word to the volume control unit 50A, and the volume control unit 50A sets the volume of the system sound based on the noise level obtained at least the speech time before the time when it received the speech time from the command utterance detection unit 30A. When the command utterance detection unit 30A detects a wake word spoken by the user from the sound picked up by the sound pickup unit 10, it detects the speaking time required to speak the wake word and sends information about the speaking time to the volume control unit 50A. When the volume control unit 50A receives information about the utterance time from the command utterance detection unit 30A, it sets the volume of the system sound based on the noise level received from the noise detection unit 20 at least the time before the utterance time from that point. During the period from when the speech time information is received from the command speech detection unit 30A to when the speech time is counted back, the sound picked up by the sound collection unit 10 includes the sound of the wake-up word spoken by the user, which may temporarily increase the noise level. For this reason, when the command utterance detection unit 30A detects a wake-up word, the volume control unit 50A changes the volume setting based on the noise level acquired before the wake-up word was uttered. That is, the volume control unit 50A sets the volume based on the noise level that does not include the voice of the wake-up word. This prevents the system sound from being output at an unnecessarily loud volume due to the user speaking the wake-up word, and prevents the volume from changing while the system sound is being output, so the user can reliably hear the system sound without any audible discomfort.

[0051] In addition, the volume control unit 50A of the information processing device 1A in this embodiment performs volume correction control in accordance with the noise level inside the vehicle, so that even in noisy environments, the system sound is output so that it can be heard at a volume close to the volume previously set by the user. This allows the user to reliably hear the system sound even in noisy environments.

[0052] Furthermore, the volume control unit 50A of the information processing device 1A according to this embodiment transmits the volume setting based on the noise level to the audio output unit 40 even when the system sound is not being output. As a result, even if the volume control unit 50A is configured to be unable to grasp the output timing of system sounds such as voice guidance, the volume setting based on the noise level is always transmitted to the audio output unit 40, so that the user can reliably hear the system sound without any audible discomfort no matter when the system sound is output.

[0053] Furthermore, the volume control unit 50A of the information processing device 1A according to this embodiment determines the volume setting of the system sound at the timing when the noise level is received from the noise detection unit 20, and transmits the volume setting to the audio output unit . When transmitting the volume setting to the audio output unit 40 in accordance with the timing at which the system sound is output, it is necessary to temporarily store the volume setting value determined based on the noise level in memory. However, since the volume control unit 50A does not need to set the volume in accordance with the timing at which the system sound is output, it is possible to eliminate the memory for storing the volume setting.

[0054] Example 3 An information processing device 1B according to this embodiment will be described with reference to FIGS. Note that components with the same reference numerals as those in the first and second embodiments have the same functions, and therefore detailed descriptions thereof will be omitted.

[0055] <Configuration of information processing device 1B> The configuration of an information processing device 1B according to this embodiment will be described with reference to FIG. The information processing device 1B is configured to include at least a sound collection unit 10, a noise detection unit 20, a command utterance detection unit 30B, a voice output unit 40, and a volume control unit 50B.

[0056] The command utterance detection unit 30B detects whether or not it is a command utterance period, which is a period from when a wake-up word is detected in the user's voice to when a command utterance from the user is detected. Specifically, the command utterance detection unit 30B detects, from the sound received from the sound collection unit 10, a wake-up word uttered by the user. Furthermore, the command utterance detection unit 30B detects a command utterance uttered by the user after detecting the wake-up word. That is, the command utterance detection unit 30B determines the period from when the wake-up word is detected until when the command utterance is detected as the command utterance period, and transmits the determination result to the noise detection unit 20. Furthermore, when the command utterance detection unit 30B detects a wake-up word, it transmits information indicating that the wake-up word has been detected to the volume control unit 50B.

[0057] The volume control unit 50B will be described with reference to FIG. If the noise level received from the noise detection unit 20 is not detected during the command utterance period, the volume control unit 50B sets the volume of the system sound based on the noise level, and if the noise level is detected during the command utterance period, the volume control unit 50B performs control not to set the volume of the audio based on the noise level. In addition, when the volume control unit 50B receives information from the command utterance detection unit 30B that a wake-up word has been detected, it sets the volume of the system sound based on the noise level obtained before a predetermined time before that point in time.

[0058] As shown in FIG. 9, the volume control unit 50B receives noise levels (N1, N2, ...) detected by the noise detection unit 20 at predetermined time intervals (e.g., every second), and determines the volume setting (S1, S2, ...) of the system voice based on the noise level values. If the noise level received from the noise detection unit 20 is not detected during the command utterance period (e.g., N1, N2, N3, N7), the volume control unit 50B transmits a volume setting value based on the noise level detected by the noise detection unit 20 to the audio output unit 40 (e.g., S1, S2, S3, S7). On the other hand, when the noise level received from the noise detection unit 20 is detected during the command utterance period (N4 to N6), the volume control unit 50B does not transmit the volume setting (S4 to S6) based on the noise level received from the noise detection unit 20 to the audio output unit 40. That is, the noise levels (N4 to N6) detected during the command utterance period include the voice of the command uttered by the user, and therefore there is a possibility that the noise levels are high. Therefore, the volume control unit 50B does not set the volume settings (S4 to S6) in the audio output unit 40 based on the noise levels (N4 to N6) acquired during this period.

[0059] As shown in Figures 9(c) to (e), when the volume control unit 50B receives information from the command utterance detection unit 30B that the user has uttered a wake-up word, it transmits to the audio output unit 40 a volume setting (S2) for the system sound based on the noise level (N2) obtained at least a predetermined time before that point in time.

[0060] Since the sound picked up from the time when the information that the wake-up word has been detected is received from the command utterance detection unit 30B to a time going back at least a predetermined time includes the voice of the wake-up word spoken by the user, the volume setting of the system sound based on the noise level before the wake-up word was spoken is transmitted to the audio output unit 40. Specifically, the noise level (N3) detected by the noise detection unit 20 is not the noise level detected during the command speaking period, but is a noise level detected that includes the sound of the wake-up word spoken by the user, so the volume control unit 50B transmits a volume setting (S2) to the audio output unit 40 based on the noise level (N2) previously obtained a predetermined time ago. More specifically, the volume control unit 50B reads from a memory (not shown) the noise level acquired a predetermined time before the time when it received information from the command utterance detection unit 30B that a wake-up word had been detected, and transmits a volume setting (S2) based on that noise level (N2) to the audio output unit 40. Here, the above-mentioned predetermined time is set to the longest utterance time value that allows the command utterance detection unit 30B to detect the utterance as a wake-up word when the user slowly utters the wake-up word. The processing flow of the volume control unit 50B will be described in detail below.

[0061] <Processing of volume control unit 50B> The processing flow of the volume control unit 50B will be described in detail with reference to FIG.

[0062] As shown in FIG. 10, the volume control unit 50B determines whether the ACC power supply (accessory power supply) of the vehicle is in an ON state (step S100). If it is determined that the ACC power supply of the vehicle is not in an ON state ("NO" in step S100), the process returns to step S100 and transitions to a standby state. On the other hand, if it is determined that the ACC power supply of the vehicle is in the ON state ("YES" in step S100), the process proceeds to step S110.

[0063] If it is determined that the ACC power supply of the vehicle is in an on state ("YES" in step S100), the volume setting to be set in the audio output unit 40 is initialized (step S110).

[0064] The volume control unit 50B acquires the detected noise level and information on whether the noise level was detected during the command utterance period from the noise detection unit 20, stores the information in a memory not shown (step S200), and proceeds to step S300. Here, the volume control unit 50B stores the information acquired from the noise detection unit 20 in a memory (not shown) in order to store the history of the detection information received from the noise detection unit 20 in chronological order. When storing information, time information or the like is added to each piece of information and stored so that it is possible to determine when the information was received from the noise detection unit 20.

[0065] The volume control unit 50B determines whether or not information indicating that a wake-up word has been detected has been received from the command utterance detection unit 30B (step S300). If it is determined that information indicating that a wake-up word has been detected has been received from command utterance detection unit 30B ("YES" in step S300), the process proceeds to step S310. On the other hand, if it is determined that information indicating that a wake-up word has been detected has not been received from command utterance detection unit 30B ("NO" in step S300), the process proceeds to step S130.

[0066] If the volume control unit 50B determines that it has received information from the command utterance detection unit 30B that a wake-up word has been detected ("YES" in step S300), it sets the volume of the system sound based on the noise level detected a predetermined time before the time when it received the information that a wake-up word has been detected (step S310), and transitions to step S130. Specifically, when the volume control unit 50B receives information that a wake-up word has been detected, it reads from a memory (not shown) the noise level last obtained before a predetermined time prior to that time, and transmits the volume setting for the system sound based on that noise level to the audio output unit 40.

[0067] The volume control unit 50B determines whether the noise level acquired from the noise detection unit 20 was detected during the command utterance period (step S130). If it is determined that the noise level acquired from the noise detection unit 20 has not been detected during the command utterance period ("NO" in step S130), the process proceeds to step S140. On the other hand, if it is determined that the noise level acquired from the noise detection unit 20 was detected during the command utterance period ("YES" in step S130), the process proceeds to step S150. In other words, if it is determined that the noise level obtained from the noise detection unit 20 is detected during the command utterance period, the noise level detected includes the voice of the user's command utterance, and therefore the volume control unit 50B performs control not to set the volume based on the noise level.

[0068] If it is determined that the noise level acquired from the noise detection unit 20 was not detected during the command utterance period ("NO" in step S130), the volume setting for the system sound is determined based on the acquired noise level, and the volume setting is sent to the audio output unit 40 (step S140). The method of determining the volume setting of the system sound based on the noise level acquired from the noise detection unit 20 is the same as in the first embodiment.

[0069] The volume control unit 50B determines whether the ACC power supply (accessory power supply) of the vehicle is in the ON state (step S150). If it is determined that the ACC power supply of the vehicle is in the ON state ("YES" in step S150), the process proceeds to step S120 and continues. On the other hand, if it is determined that the ACC power supply of the vehicle is not in an ON state ("NO" in step S150), the process is ended.

[0070] When the command utterance detection unit 30B of the information processing device 1B in this embodiment detects a wake-up word, it transmits information that the wake-up word has been detected to the volume control unit 50B, and the volume control unit 50B sets the volume of the audio (system sound) based on the noise level obtained at least a predetermined time before the time when it received the information that the wake-up word has been detected from the command utterance detection unit 30B. The predetermined time is set to the longest utterance time during which the command utterance detection unit 30B can detect the wake-up word. When the command utterance detection unit 30B detects a wake word uttered by the user from the sound collected by the sound collection unit 10, the command utterance detection unit 30B transmits information indicating that a wake-up word has been detected to the volume control unit 50B. During the period from the time when information that a wake-up word has been detected is received from the command utterance detection unit 30B to a time going back a predetermined time, the sound picked up by the sound collection unit 10 includes the voice of the wake-up word spoken by the user, which may result in a high noise level. Therefore, when the volume control unit 50B receives information from the command utterance detection unit 30B that a wake-up word has been detected, it sets the volume of the system sound based on the noise level detected by the noise detection unit 20 that last detected the wake-up word at least a predetermined time before that point in time. Since the above-mentioned predetermined time is set to the longest speech time during which the command speech detection unit 30B can detect a wake-up word, the volume of the system sound can be set based on the noise level detected from the voice spoken by the user that does not include the wake-up word. This prevents the system sound from being output at an unnecessarily loud volume due to the user speaking the wake-up word, allowing the user to reliably hear the system sound without any discomfort to the ear.

[0071] In addition, the volume control unit 50B of the information processing device 1B according to this embodiment performs volume correction control in accordance with the noise level inside the vehicle, so that even in noisy environments, the system sound is output at a volume close to the volume previously set by the user. This allows the user to reliably hear the system sound even in noisy environments.

[0072] Furthermore, the volume control unit 50B of the information processing device 1B according to this embodiment transmits the volume setting based on the noise level to the audio output unit 40 even when the system sound is not being output. As a result, even if the volume control unit 50B is configured to be unable to grasp the output timing of system sounds such as voice guidance, the volume setting based on the noise level is always transmitted to the audio output unit 40, so that the user can reliably hear the system sound without any audible discomfort no matter when the system sound is output.

[0073] Furthermore, the volume control unit 50B of the information processing device 1B according to this embodiment determines the volume setting of the system sound at the timing when the noise level is received from the noise detection unit 20, and transmits the volume setting to the audio output unit 40. When transmitting the volume setting to the audio output unit 40 in accordance with the timing at which the system sound is output, it is necessary to temporarily store the volume setting value determined based on the noise level in memory. However, since volume control unit 50B does not need to set the volume in accordance with the timing at which the system sound is output, it is possible to eliminate the memory for storing the volume setting.

[0074] <Other Examples> In the above-described information processing device 1, the detection result of the command utterance detection unit 30 is transmitted to the noise detection unit 20, but it may be transmitted to the volume control unit . In other words, in the information processing device 1, the noise detection unit 20 determines whether a noise level was detected during the command utterance period, but the volume control unit 50 may receive the detection result from the command utterance detection unit 30 and determine whether the acquired noise level was detected during the command utterance period. Specifically, the volume control unit 50 determines that during the period in which it receives information from the command utterance detection unit 30 that it is a command utterance period, the noise level received from the noise detection unit 20 includes the user's command utterance voice. In addition, immediately after the command utterance period ends and the volume control unit 50 receives information from the command utterance detection unit 30 that it is no longer a command utterance period, the volume control unit 50 determines that the noise level received from the noise detection unit 20 includes the user's command utterance voice. If it is determined that the noise level received from the noise detection unit 20 was not detected during the command utterance period, the volume control unit 50 transmits a volume setting based on the noise level to the audio output unit 40. On the other hand, if it is determined that the noise level received from the noise detection unit 20 was detected during the command utterance period, the volume setting based on the noise level is not transmitted to the audio output unit 40. This eliminates volume changes during voice output due to user command utterances, and voice output with unnecessarily increased volume, so the user can reliably hear what is being said without any sense of discomfort to the ear.

[0075] In the information processing devices 1, 1A, and 1B, when the server executes the voice recognition process for executing the voice assistant function, the process of the command utterance detection units 30, 30A, and 30B may be executed by the server. Specifically, the audio picked up by the audio pickup unit 10 is transferred to a server connected to the Internet line, and the server executes the processing of the command utterance detection units 30, 30A, and 30B, which is to detect whether or not it is a command utterance period and to detect the time when the wake-up word is spoken. The detection results detected by the server are transmitted to the information processing devices 1, 1A, and 1B via the Internet line. The processing of the command utterance detection units 30, 30A, and 30B involves recognizing wake-up words and command utterances from the user's speech, and therefore requires resources such as an SoC (System on Chip) capable of extremely high-speed computational processing, as well as memory for such processing. In addition, the processing of the voice assistant function requires an SoC capable of extremely high-speed computational processing, as well as resources such as memory for such processing, in order to recognize the device's operation instructions and information search instructions from the user's spoken voice and execute those instructions. In other words, similar resources are required to execute the processing of the command utterance detection units 30, 30A, 30B and the voice assistant function. Furthermore, the processing of the command utterance detection units 30, 30A, 30B and the voice assistant function is performed based on the sound collected by the sound collection unit 10. In other words, when the voice assistant function is processed on the server, the circuit of the command utterance detection unit 30 in the information processing device 1, 1A, 1B is deleted, and the voice picked up by the sound pickup unit 10 is sent to the server, which then executes the processing of the command utterance detection units 30, 30A, 30B. This makes it possible to eliminate the circuits of the command utterance detection units 30, 30A, and 30B, thereby reducing the cost of the information processing devices 1, 1A, and 1B. Furthermore, since it is possible to eliminate the need for an SoC that performs high-speed arithmetic processing, a memory, and the like, it is possible to reduce the power consumption of the information processing devices 1, 1A, and 1B. Furthermore, since the circuits of the command utterance detection units 30, 30A, and 30B can be eliminated, the weight of the information processing devices 1, 1A, and 1B can be reduced.

[0076] Although an embodiment of the present invention has been described above in detail with reference to the drawings, the specific configuration is not limited to this embodiment, and includes designs within the scope of the gist of the present invention. [Explanation of symbols]

[0077] 1. Information processing device 1A: Information processing device 1B: Information processing device 10; Sound pickup section 20: Noise detection unit 30: Command utterance detection unit 30A: Command utterance detection unit 30B: Command utterance detection unit 40: Audio output section 50;Volume control 50A;Volume control 50B: Volume control unit

Claims

1. a noise detection unit that detects a noise level; a command utterance detection unit that detects whether or not a command utterance period is occurring, which is a period from when a wake-up word is detected in a user's voice to when a command utterance from the user is detected; a volume control unit that sets a volume of a voice to be output to the user based on the noise level when the detection of the noise level is not during the command utterance period, and that sets the volume of the voice based on a noise level acquired at a time prior to a predetermined time before the detection of the wake-up word when the detection of the noise level is during the command utterance period; An information processing device comprising:

2. The information processing device according to claim 1, characterized in that, when the noise level detection is during the command utterance period, the volume control unit sets the volume of the audio based on the noise level obtained before a point that is the predetermined time before the time when the wake-up word was detected, which is the speaking time required for the user to utter the wake-up word.

3. The information processing device according to claim 1, characterized in that, when the noise level detection is during the command utterance period, the volume control unit sets the volume of the audio based on the noise level acquired before the point in time going back the predetermined time from the point in time when the wake-up word was detected, which is the longest utterance time at which the command utterance detection unit can detect the wake-up word.

4. 4. The information processing apparatus according to claim 1, wherein the voice is a voice guidance sound for guiding the user along a route to a destination set by the user.

5. 4. The information processing apparatus according to claim 1, wherein the voice is a voice guidance sound including recommendation information for the user.

6. An information processing method for an information processing device including a noise detection unit, a command utterance detection unit, and a volume control unit, a first step in which the noise detection unit detects a noise level; a second step of detecting whether or not the command utterance detection unit is in a command utterance period, which is a period from when a wake-up word is detected from a user's speech to when a command utterance from the user is detected; a third step in which the volume control unit sets the volume of the sound to be output to the user based on the noise level when the detection of the noise level is not during the command utterance period, and sets the volume of the sound based on the noise level acquired at a time prior to a predetermined time before the detection of the wake-up word when the detection of the noise level is during the command utterance period; An information processing method comprising:

7. A program for causing a computer to execute an information processing method for an information processing device including a noise detection unit, a command utterance detection unit, and a volume control unit, a first step in which the noise detection unit detects a noise level; a second step of detecting whether or not the command utterance detection unit is in a command utterance period, which is a period from when a wake-up word is detected from a user's speech to when a command utterance from the user is detected; a third step in which the volume control unit sets the volume of the sound to be output to the user based on the noise level when the detection of the noise level is not during the command utterance period, and sets the volume of the sound based on the noise level acquired at a time prior to a predetermined time before the detection of the wake-up word when the detection of the noise level is during the command utterance period; A program for causing a computer to execute an information processing method comprising:

8. A computer-readable non-transitory recording medium having a program recorded thereon for causing a computer to execute an information processing method for an information processing device including a noise detection unit, a command utterance detection unit, and a volume control unit, a first step in which the noise detection unit detects a noise level; a second step of detecting whether or not the command utterance detection unit is in a command utterance period, which is a period from when a wake-up word is detected from a user's speech to when a command utterance from the user is detected; a third step in which the volume control unit sets the volume of the sound to be output to the user based on the noise level when the detection of the noise level is not during the command utterance period, and sets the volume of the sound based on the noise level acquired at a time prior to a predetermined time before the detection of the wake-up word when the detection of the noise level is during the command utterance period; A non-transitory recording medium having a program recorded thereon for causing a computer to execute an information processing method comprising the steps of:

Images