Speech recognition method, speech wakeup apparatus, speech recognition apparatus, and terminal
Embodiments of the present invention provide a speech recognition method and a terminal. The method includes: listening, by a speech wakeup apparatus, to speech information in a surrounding environment; when determining that the speech information obtained by listening matches a speech wakeup model, buffering, by the speech wakeup apparatus, speech information, of first preset duration, obtained by listening, and sending a trigger signal for triggering enabling of a speech recognition apparatus, where the trigger signal is used to instruct the speech recognition apparatus to read and recognize the speech information buffered by the speech wakeup apparatus; and recognizing first speech information buffered by the speech wakeup apparatus and the second speech information obtained by listening, to obtain a recognition result.
1. A speech control method, applied to a terminal that includes a speech wakeup apparatus and a speech recognition apparatus, the method comprising:
listening, by the speech wakeup apparatus, for first speech information in a surrounding environment, wherein the first speech information comprises wakeup information and a first portion of a command word, wherein the wakeup information is used to enable the speech recognition apparatus;
generating, by the speech wakeup apparatus, a trigger signal based on determining that the wakeup information matches a speech wakeup model;
sending, by the speech wakeup apparatus, the trigger signal to the speech recognition apparatus;
enabling, by the speech wakeup apparatus, the speech recognition apparatus according to the trigger signal;
listening, by the speech recognition apparatus, for second speech information, wherein the second speech information comprises a second portion of the command word; and
obtaining, by the speech recognition apparatus, speech instruction information according to the first speech information and the second speech information, wherein the speech instruction information matches the command word, and the command word comprises the first portion of the command word and the second portion of the command word.
2. The method according to claim 1 , wherein the determining that the wakeup information matches the speech wakeup model comprises:
in a case that the wakeup information matches predetermined wakeup speech information, determining that the wakeup information matches the speech wakeup model.
3. The method according to claim 1 , wherein the determining that the wakeup information matches the speech wakeup model comprises:
in a case that the wakeup information matches predetermined wakeup speech information, extracting a voiceprint feature in the wakeup information, and in a case that the extracted voiceprint feature matches a predetermined voiceprint feature, determining that the wakeup information matches the speech wakeup model.
4. The method according to claim 3 , wherein the voiceprint feature includes one or more of the following features:
a pitch contour, a linear prediction coefficient, a spectral envelope parameter, a harmonic energy ratio, a resonant peak frequency and its bandwidth, a cepstrum, or a Mel-frequency cepstrum coefficient.
5. The method according to claim 1 , wherein the obtaining, by the speech recognition apparatus, the speech instruction information according to the first speech information and the second speech information comprises:
obtaining, by the speech recognition apparatus, a recognition result according to the first speech information and the second speech information, wherein the recognition result comprises command word information; and
obtaining, by the speech recognition apparatus, the speech instruction information that matches the recognition result by matching between the obtained recognition result and pre-stored speech instruction information.
6. The method according to claim 1 ,
wherein the wakeup information is listened in a first period by the speech wakeup apparatus, and the first portion of the command word is listened in a second period by the speech wakeup apparatus; and
wherein the second speech information is listened in a third period by the speech recognition apparatus.
7. The method according to claim 1 , wherein the listening, by the speech wakeup apparatus, for the first speech information in the surrounding environment comprises:
listening for the first speech information in the surrounding environment in a standby state; or
listening for the first speech information in the surrounding environment in a non-standby state; or
listening for the first speech information in the surrounding environment in a screen-locked state.
8. The method according to claim 1 , further comprising:
controlling, by the speech recognition apparatus, execution of an operation corresponding to the speech instruction information.
9. The method according to claim 1 , wherein the first portion of the command word is listened by the speech wakeup apparatus before the speech recognition apparatus is enabled.
10. A terminal, comprising:
a speech wakeup apparatus; and
a speech recognition apparatus;
wherein the speech wakeup apparatus is configured to:
listen for first speech information in a surrounding environment, wherein the first speech information comprises a wakeup information and a first portion of a command word, wherein the wakeup information is used to enable the speech recognition apparatus;
generate a trigger signal based on determining that the wakeup information matches a speech wakeup model;
send the trigger signal to the speech recognition apparatus; and
enable the speech recognition apparatus according to the trigger signal; and
wherein the speech recognition apparatus is configured to:
listen for second speech information, wherein the second speech information comprises a second portion of the command word; and
obtain speech instruction information according to the first speech information and the second speech information, wherein the speech instruction information matches the command word, and the command word comprises the first portion of the command word and the second portion of the command word.
11. The terminal according to claim 10 , wherein the speech wakeup apparatus is configured to:
in a case that the wakeup information matches predetermined wakeup speech information, extract a voiceprint feature in the wakeup information, and in a case that the extracted voiceprint feature matches a predetermined voiceprint feature, determine that the wakeup information matches a speech wakeup model.
12. The terminal according to claim 11 , wherein the voiceprint feature includes one or more of the following features:
a pitch contour, a linear prediction coefficient, a spectral envelope parameter, a harmonic energy ratio, a resonant peak frequency and its bandwidth, a cepstrum, or a Mel-frequency cepstrum coefficient.
13. The terminal according to claim 10 , wherein the speech recognition apparatus is configured to:
obtain a recognition result according to the first speech information and the second speech information, wherein the recognition result comprises command word information; and
obtain the speech instruction information that matches the recognition result by matching between the obtained recognition result and pre-stored speech instruction information.
14. The terminal according to claim 10 ,
wherein the wakeup information is listened in a first period by the speech wakeup apparatus, and the first portion of the command word is listened in a second period by the speech wakeup apparatus; and
wherein the second speech information is listened in a third period by the speech recognition apparatus.
15. The terminal according to claim 10 , wherein the speech wakeup apparatus is configured to:
listen for the first speech information in the surrounding environment in a standby state; or
listen for the first speech information in the surrounding environment in a non-standby state; or
listen for the first speech information in the surrounding environment in a screen-locked state.
16. The terminal according to claim 10 , wherein the speech recognition apparatus is configured to:
automatically disable when determining that speech information is not received again within a preset duration after enabling the speech recognition apparatus.
17. The terminal according to claim 10 , further comprising:
a processor;
wherein the speech recognition apparatus is also configured to send an execution instruction that matches the speech instruction information to the processor; and
the processor is configured to execute an operation corresponding to the execution instruction.
18. The terminal according to claim 10 , wherein the speech wakeup apparatus is a digital signal processor (DSP), and the speech recognition apparatus is an application processor (AP).
19. The terminal according to claim 10 , wherein the first portion of the command word is listened by the speech wakeup apparatus before the speech recognition apparatus is enabled.