Information processing apparatus, information processing method, program and recording medium
An information processing apparatus controls volume so that a user can reliably hear voice guidance and other audio without auditory discomfort and with certainty. A noise detector detects a noise level in a vehicle interior collected in a sound collector, and a command speech detector detects whether it is a command speech period, which is a period from detection of a wake-up word from the user's speech sound to the detection of user's command speech. A volume controller, which sets volume of sounds to be output to the user based on the noise level when the noise level is not detected during the command speech period, and does not set the volume of the sounds based on the noise level when the noise level is detected during the command speech period.
1 . An information processing apparatus comprising:
a noise detection circuit that divides an audio signal into a plurality of time segments and calculates an average noise level for each of the plurality of time segments to detect noise level;
a command speech detection circuit that detects whether it is a command speech period, which is a period of time between detection of a wake-up word from a user's speech sound and the detection of user's command speech; and
a volume controller including one or more processors,
wherein the volume controller
adjusts an audio output level of a voice guidance sound based on the average noise level detected at a time other than during the command speech period;
adjusts the audio output level of the voice guidance sound based on the average noise level acquired before a point backward in a predetermined time from the point of when the wake-up word has been detected, during the command speech period; and
the information processing apparatus outputs the voice guidance sound at the adjusted audio output level.
2 . The information processing apparatus according to claim 1 ,
wherein the volume controller, when the noise level is detected during the command speech period, adjusts the volume of the voice guidance sound based on the average noise level acquired before the point backward in speech time taken by the user to speak the wake-up word, which is the point backward in the predetermined time from the point of when the wake-up word has been detected.
3 . The information processing apparatus according to claim 2 ,
wherein the voice guidance sound guides a route to a destination set by the user.
4 . The information processing apparatus according to claim 2 ,
wherein the voice guidance sound includes recommendation information for the user.
5 . The information processing apparatus according to claim 1 ,
wherein the volume controller, when the noise level is detected during the command speech period, adjusts the volume of the voice guidance sound based on the average noise level acquired before the point backward in maximum speech time during which the command speech-detection circuit can detect the wake-up word, which is the point backward in the predetermined time, from the point of when the wake-up word has been detected.
6 . The information processing apparatus according to claim 5 ,
wherein the voice guidance sound guides a route to a destination set by the user.
7 . The information processing apparatus according to claim 5 ,
wherein the voice guidance sound includes recommendation information for the user.
8 . The information processing apparatus according to claim 1 , wherein the voice guidance sound guides a route to a destination set by the user.
9 . The information processing apparatus according to claim 1 ,
wherein the voice guidance sound includes recommendation information for the user.
10 . An information processing method comprising:
dividing, by an information processing apparatus, an audio signal into a plurality of time segments;
calculating, by the information processing apparatus, an average noise level for each of the plurality of time segments to detect a noise level;
detecting, by the information processing apparatus, whether it is a command speech period, which is a period of time between detection of a wake-up word from a user's speech sound and the detection of user's command speech;
adjusting, by a volume controller including one or more processors of the information processing apparatus, an audio output level of a voice guidance sound based on the average noise level detected at a time other than during the command speech period;
adjusting, by the volume controller, the audio output level of the voice guidance sound based on the average noise level acquired before a point backward in a predetermined time from the point of when the wake-up word has been detected, during the command speech period; and
outputting, by the information processing apparatus, the voice guidance sound at the adjusted audio output level.
11 . A non-transitory computer-readable recording medium storing a program causing a computer to execute
dividing an audio signal into a plurality of time segments;
calculating an average noise level for each of the plurality of time segments to detect a noise level;
detecting whether it is a command speech period, which is a period of time between detection of a wake-up word from a user's speech sound and the detection of user's command speech;
adjusting an audio output level of a voice guidance sound based on the average noise level at a time other than during the command speech period;
adjusting the audio output level of the voice guidance sound based on the average noise level acquired before a point backward in a predetermined time from the point of when the wake-up word has been detected, during the command speech period; and
outputting the voice guidance sound at the adjusted audio output level.