Natural-language processing across multiple languages
A method includes obtaining a query in a base language and translating the query to generate one or more translated queries each in a respective target language. The method also includes searching one or more sets of electronic files based on the one or more translated queries to generate target-language search results, where each translated query is used to search one or more electronic files that include content in the respective target language of the translated query. The method also includes, based on the target-language search results, scheduling one or more electronic files of the one or more sets of electronic files for at least partial translation to the base language.
1. A device comprising:
one or more processors configured to:
obtain a query in a base language and a conceptual tag associated with the query;
translate the query, based on the conceptual tag, to generate a translated query in a target language, wherein the conceptual tag is indicative of a context relevant to translation of the query, wherein the query is translated using a first context-specific dictionary responsive to the conceptual tag being a first particular conceptual tag, and wherein the query is translated using a second context-specific dictionary distinct from the first context-specific dictionary responsive to the conceptual tag being a second particular conceptual tag distinct from the first particular conceptual tag;
search a first set of electronic files based on the translated query to generate target-language search results, wherein the first set of electronic files include content in the target language;
translate portions of one or more electronic files listed in the target-language search results to the base language to generate a base-language search results listing; and
generate an output including at least a portion of the base-language search results listing.
2. The device of claim 1 , wherein the one or more processors are further configured to:
receive input indicating a level of pertinence of a particular base-language search result of the base-language search results listing; and
based on the input, schedule further translation of an electronic file corresponding to the particular base-language search result.
3. The device of claim 1 , wherein the one or more processors are further configured to save the query, translated query, or both, to a persistent query list for use in searching one or more additional electronic files that become available after the searching of the one or more sets of electronic files.
4. The device of claim 1 , wherein the one or more processors are further configured to, before translating the query to generate the translated query, select the target language based on configurable query parameters that are based on a data acquisition requirement.
5. The device of claim 1 , wherein the output includes base-language search results snippets based on the translated portions of the one or more electronic files, and wherein the one or more processors are further configured to:
receive user input indicating a level of pertinence of a particular base-language search results snippet; and
determine whether to update a translation schedule based on the user input.
6. The device of claim 1 , wherein the output includes base-language search results snippets based on the translated portions of the one or more electronic files, and wherein the one or more processors are further configured to:
receive user input indicating a level of pertinence of a particular base-language search results snippet; and
determine whether to modify the translated query based on the user input.
7. The device of claim 1 , wherein the output includes base-language search results snippets based on the translated portions of the one or more electronic files, and wherein the one or more processors are further configured to:
receive user input indicating a level of pertinence of a particular base-language search results snippet;
generate or update a machine-learning model based on the user input; and
use the machine-learning model to filter search results, to expand a search query, to assign label data to the one or more electronic files or portions of the one or more electronic files, or a combination thereof.
8. The device of claim 1 , wherein a particular electronic file of the first set of electronic files includes audio data representing speech in the target language, wherein the one or more processors are further configured to:
after translating the query to generate the translated query, generate an audio signature representing at least a portion of the translated query spoken in the target language, and wherein searching the first sets of electronic files includes comparing the audio signature to the audio data.
9. The device of claim 1 , wherein the one or more processors are further configured to:
obtain a second conceptual tag associated with the query;
translate the query, based on the second conceptual tag, to generate a second translated query in the target language, wherein the second conceptual tag is indicative of a second context relevant to translation of the query, and wherein the second translated query is different from the translated query; and
search at least one of the first set of electronic files or a second set of electronic files based on the second translated query to generate the target-language search results.
10. The device of claim 1 , wherein the one or more processors are further configured to:
obtain a second conceptual tag associated with the query; and
based on the second conceptual tag, search a second set of electronic files based on the translated query.
11. A method comprising:
obtaining a query in a base language and a conceptual tag associated with the query;
translating the query, based on the conceptual tag, to generate a translated query in a target language, wherein the conceptual tag is indicative of a context relevant to translation of the query, wherein the query is translated using a first context-specific dictionary responsive to the conceptual tag being a first particular conceptual tag, and wherein the query is translated using a second context-specific dictionary distinct from the first context-specific dictionary responsive to the conceptual tag being a second particular conceptual tag distinct from the first particular conceptual tag;
searching a first set of electronic files based on the translated query to generate target-language search results, wherein the first set of electronic files includes content in the target language;
translating portions of one or more electronic files listed in the target-language search results to the base language to generate a base-language search results listing; and
generating an output including at least a portion of the base-language search results listing.
12. The method of claim 11 , further comprising:
receiving input indicating a level of pertinence of a particular base-language search result of the base-language search results listing; and
based on the input, scheduling further translation of an electronic file corresponding to the particular base-language search result.
13. The method of claim 11 , further comprising saving the query to a persistent query list for use in searching one or more additional electronic files that become available after the searching of the one or more sets of electronic files.
14. The method of claim 11 , further comprising saving the translated query to a persistent query list for use in searching one or more additional electronic files that become available after the searching of the one or more sets of electronic files.
15. The method of claim 11 , wherein the output includes base-language search results snippets based on the translated portions of the one or more electronic files, and further comprising:
receiving user input indicating a level of pertinence of a particular base-language search results snippet;
generating or updating a machine-learning model based on the user input; and
using the machine-learning model to filter search results, to expand a search query, or to assign label data to the one or more electronic files or portions of the one or more electronic files, or a combination thereof.
16. The method of claim 11 , further comprising:
obtaining a second conceptual tag associated with the query;
translating the query, based on the second conceptual tag, to generate a second translated query in the target language, wherein the second conceptual tag is indicative of a second context relevant to translation of the query, and wherein the second translated query is different from the translated query; and
searching at least one of the first set of electronic files or a second set of electronic files based on the second translated query to generate the target-language search results.
17. A non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to:
obtain a query in a base language and a conceptual tag associated with the query;
translate the query, based on the conceptual tag, to generate a translated query in a target language, wherein the conceptual tag is indicative of a context relevant to translation of the query, wherein the query is translated using a first context-specific dictionary responsive to the conceptual tag being a first particular conceptual tag, and wherein the query is translated using a second context-specific dictionary distinct from the first context-specific dictionary responsive to the conceptual tag being a second particular conceptual tag distinct from the first particular conceptual tag;
search a first set of electronic files based on the translated query to generate target-language search results, wherein the first set of electronic files include content in the target language;
translate portions of one or more electronic files listed in the target-language search results to the base language to generate a base-language search results listing; and
generate an output including at least a portion of the base-language search results listing.
18. The non-transitory computer-readable medium of claim 17 , wherein execution of the instructions further causes the one or more processors to:
receive input indicating a level of pertinence of a particular base-language search result of the base-language search results listing; and
based on the input, schedule further translation of an electronic file corresponding to the particular base-language search result.
19. The non-transitory computer-readable medium of claim 17 , wherein the output includes base-language search results snippets based on the translated portions of the one or more electronic files, and wherein execution of the instructions further causes the one or more processors to:
receive user input indicating a level of pertinence of a particular base-language search results snippet; and
generate or update a machine-learning model based on the user input.
20. The non-transitory computer-readable medium of claim 17 , wherein execution of the instructions further causes the one or more processors to:
obtain a second conceptual tag associated with the query;
translate the query, based on the second conceptual tag, to generate a second translated query in the target language, wherein the second conceptual tag is indicative of a second context relevant to translation of the query, and wherein the second translated query is different from the translated query; and
search at least one of the first set of electronic files or a second set of electronic files based on the second translated query to generate the target-language search results.