Speech endpointing based on word comparisons
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for speech endpointing based on word comparisons are described. In one aspect, a method includes the actions of obtaining a transcription of an utterance. The actions further include determining, as a first value, a quantity of text samples in a collection of text samples that (i) include terms that match the transcription, and (ii) do not include any additional terms. The actions further include determining, as a second value, a quantity of text samples in the collection of text samples that (i) include terms that match the transcription, and (ii) include one or more additional terms. The actions further include classifying the utterance as a likely incomplete utterance or not a likely incomplete utterance based at least on comparing the first value and the second value.
1. A computer-implemented method comprising:
obtaining a transcription of an utterance;
determining a first quantity of text samples in a collection of text samples that (i) include terms that match the transcription, and (ii) do not include any additional terms;
determining a second quantity of text samples in the collection of text samples that (i) include terms that match the transcription, and (ii) include one or more additional terms;
comparing the first quantity and the second quantity;
classifying the utterance as a likely incomplete utterance or not a likely incomplete utterance based at least on comparing the first quantity and the second quantity; and
maintaining a microphone in an active state to receive an additional utterance based on classifying the utterance as a likely incomplete, or deactivating the microphone based on classifying the utterance as not a likely incomplete utterance.
2. The method of claim 1 , wherein determining a first quantity of text samples in a collection of text samples that (i) include terms that match the transcription, and (ii) do not include any additional terms comprises:
determining that, in each text sample, that terms that match the transcription occur in a same order as in the transcription.
3. The method of claim 1 , wherein determining a second quantity of text samples in the collection of text samples that (i) include terms that match the transcription, and (ii) include one or more additional terms comprises:
determining that, in each text sample, the terms that match the transcription occur at a prefix of each text sample.
4. The method of claim 1 , wherein:
comparing the first quantity and the second quantity comprises:
determining a ratio of the first quantity to the second quantity; and
determining that the ratio satisfies a threshold ratio; and
classifying the utterance as a likely incomplete utterance or not a likely incomplete utterance based at least on comparing the first quantity and the second quantity comprises:
based on determining that the ratio satisfies the threshold ratio, classifying the utterance as a likely incomplete utterance.
5. The method of claim 1 , wherein:
classifying the utterance as a likely incomplete utterance or not a likely incomplete utterance based at least on comparing the first quantity and the second quantity comprises classifying the utterance as a likely incomplete utterance; and
maintaining a microphone in an active state to receive an additional utterance based on classifying the utterance as a likely incomplete, or deactivating the microphone based on classifying the utterance as not a likely incomplete utterance comprises based on classifying the utterance as a likely incomplete utterance, maintaining a microphone in an active state to receive an additional utterance.
6. The method of claim 1 , wherein:
classifying the utterance as a likely incomplete utterance or not a likely incomplete utterance based at least on comparing the first quantity and the second quantity comprises classifying the utterance as not a likely incomplete utterance; and
maintaining a microphone in an active state to receive an additional utterance based on classifying the utterance as a likely incomplete, or deactivating the microphone based on classifying the utterance as not a likely incomplete utterance comprises based on classifying the utterance as not a likely incomplete utterance, deactivating a microphone.
7. The method of claim 1 , comprising:
receiving data indicating that the utterance is complete;
wherein classifying the utterance as a likely incomplete utterance or not a likely incomplete utterance based at least on comparing the first quantity and the second quantity comprises classifying the utterance as a likely incomplete utterance; and
based on classifying the utterance as a likely incomplete utterance, overriding the data indicating that the utterance is complete.
8. The method of claim 1 , wherein the transcription of the utterance is a search query and the collection of text samples is a collection of search queries.
9. A system comprising:
one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:
obtaining a transcription of an utterance;
determining a first quantity of text samples in a collection of text samples that (i) include terms that match the transcription, and (ii) do not include any additional terms;
determining second quantity of text samples in the collection of text samples that (i) include terms that match the transcription, and (ii) include one or more additional terms;
comparing the first quantity and the second quantity;
classifying the utterance as a likely incomplete utterance or not a likely incomplete utterance based at least on comparing the first quantity and the second quantity; and
maintaining a microphone in an active state to receive an additional utterance based on classifying the utterance as a likely incomplete, or deactivating a microphone based on classifying the utterance as not a likely incomplete utterance.
10. The system of claim 9 , wherein determining a first quantity of text samples in a collection of text samples that (i) include terms that match the transcription, and (ii) do not include any additional terms comprises:
determining that, in each text sample, that terms that match the transcription occur in a same order as in the transcription.
11. The system of claim 9 , wherein determining a second quantity of text samples in the collection of text samples that (i) include terms that match the transcription, and (ii) include one or more additional terms comprises:
determining that, in each text sample, the terms that match the transcription occur at a prefix of each text sample.
12. The system of claim 9 , wherein:
comparing the first quantity and the second quantity comprises:
determining a ratio of the first quantity to the second quantity; and
determining that the ratio satisfies a threshold ratio; and
classifying the utterance as a likely incomplete utterance or not a likely incomplete utterance based at least on comparing the first quantity and the second quantity comprises:
based on determining that the ratio satisfies the threshold ratio, classifying the utterance as a likely incomplete utterance.
13. The system of claim 9 , wherein:
classifying the utterance as a likely incomplete utterance or not a likely incomplete utterance based at least on comparing the first quantity and the second quantity comprises classifying the utterance as a likely incomplete utterance; and
maintaining a microphone in an active state to receive an additional utterance based on classifying the utterance as a likely incomplete, or deactivating the microphone based on classifying the utterance as not a likely incomplete utterance comprises based on classifying the utterance as a likely incomplete utterance, maintaining a microphone in an active state to receive an additional utterance.
14. The system of claim 9 , wherein:
classifying the utterance as a likely incomplete utterance or not a likely incomplete utterance based at least on comparing the first quantity and the second quantity comprises classifying the utterance as not a likely incomplete utterance; and
maintaining a microphone in an active state to receive an additional utterance based on classifying the utterance as a likely incomplete, or deactivating the microphone based on classifying the utterance as not a likely incomplete utterance comprises based on classifying the utterance as not a likely incomplete utterance, deactivating a microphone.
15. The system of claim 9 , the operations further comprising:
receiving data indicating that the utterance is complete;
wherein classifying the utterance as a likely incomplete utterance or not a likely incomplete utterance based at least on comparing the first quantity and the second quantity comprises classifying the utterance as a likely incomplete utterance; and
based on classifying the utterance as a likely incomplete utterance, overriding the data indicating that the utterance is complete.
16. A non-transitory computer-readable medium storing software comprising instructions executable by one or more computers which, upon such execution, cause the one or more computers to perform operations comprising:
obtaining a transcription of an utterance;
determining a first quantity of text samples in a collection of text samples that (i) include terms that match the transcription, and (ii) do not include any additional terms;
determining a second quantity of text samples in the collection of text samples that (i) include terms that match the transcription, and (ii) include one or more additional terms;
comparing the first quantity and the second quantity;
classifying the utterance as a likely incomplete utterance or not a likely incomplete utterance based at least on comparing the first quantity and the second quantity; and
maintaining a microphone in an active state to receive an additional utterance based on classifying the utterance as a likely incomplete, or deactivating the microphone based on classifying the utterance as not a likely incomplete utterance.
17. The medium of claim 16 , wherein determining a first quantity of text samples in a collection of text samples that (i) include terms that match the transcription, and (ii) do not include any additional terms comprises:
determining that, in each text sample, the terms that match the transcription occur at a prefix of each text sample.
18. The medium of claim 16 , wherein:
comparing the first quantity and the second quantity comprises:
determining a ratio of the first quantity to the second quantity; and
determining that the ratio satisfies a threshold ratio; and
classifying the utterance as a likely incomplete utterance or not a likely incomplete utterance based at least on comparing the first quantity and the second quantity comprises:
based on determining that the ratio satisfies the threshold ratio, classifying the utterance as a likely incomplete utterance.
19. The medium of claim 16 , wherein:
classifying the utterance as a likely incomplete utterance or not a likely incomplete utterance based at least on comparing the first quantity and the second quantity comprises classifying the utterance as a likely incomplete utterance; and
maintaining a microphone in an active state to receive an additional utterance based on classifying the utterance as a likely incomplete, or deactivating the microphone based on classifying the utterance as not a likely incomplete utterance comprises based on classifying the utterance as a likely incomplete utterance, maintaining a microphone in an active state to receive an additional utterance.
20. The medium of claim 16 , the operations further comprising:
receiving data indicating that the utterance is complete;
wherein classifying the utterance as a likely incomplete utterance or not a likely incomplete utterance based at least on comparing the first quantity and the second quantity comprises classifying the utterance as a likely incomplete utterance; and
based on classifying the utterance as a likely incomplete utterance, overriding the data indicating that the utterance is complete.