100 receives the audio signal 122 at a microphone of the first electronic device 100, and the first electronic device 100 can generate the first text representation 120 based on audible signal data that corresponds to the audio signal 122.

In some implementations, the second electronic device 150 determines language context information 130 that is associated with the second electronic device 150. In some implementations, the language context information 130 indicates a primary language associated with the second electronic device 150 (e.g., a default language of the second electronic device 150 or a preferred language of the user 50). In some implementations, the language context information 130 indicates a location that is associated with the second electronic device 150 (e.g., a home location of the second electronic device 150). The second electronic device 150 may determine the language context information 130 based on information available to the second electronic device 150. For example, in some implementations, the second electronic device 150 determines the language context information 130 based on a device profile that is associated with the second electronic device 150. The device profile may associate the second electronic device 150 with a geographic region, such as a country where a particular language is spoken. In some implementations, the second electronic device 150 determines the language context information 130 based on a user profile that is associated with the user 50. For example, the user profile may indicate a preferred language of the user 50. As another example, the user profile may associate the user 50 with a geographic region, such as a country where a particular language is spoken. In some implementations, the second electronic device 150 determines the language context information 130 based on GPS information, which may be mapped to a geographic region where a particular language is spoken. The second electronic device 150 may determine the language context information 130 based on a network identifier, such as a cell identifier or an SSID, which may be associated with a geographic region where a particular language is spoken.

In some implementations, the second electronic device 150 determines the language context information 130 based on the audio signal 122. For example, the second electronic device 150 may detect a spoken language associated with a spoken phrase based on the audio signal 122. The second electronic device 150 may determine the language context information 130 based on the spoken language.

In some implementations, the second electronic device 150 determines the language context information 130 based on the first text representation 120. For example, the second electronic device 150 may analyze the first text representation 120 to determine a language represented in the first text representation 120. The second electronic device 150 may determine the language context information 130 based on the language.

In some implementations, the second electronic device 150 receives the language context information 130 from a database 132. The database 132 may store data associating electronic devices, including the second electronic device 150, with respective geographic regions. For example, the database 132 may reside at a network element (e.g., a cellular network element such as a Home Location Register (HLR)) that stores a home location of the second electronic device 150.

The language context information 130 may include geographic information. The geographic information may include a country identifier. In some implementations, the geographic information includes a political subdivision of a country, e.g., a province or a state. In some implementations, the language context information 130 includes a language identifier. For example, the language context information 130 may specify that the user resides in Canada and speaks English or French.

In some implementations, the language context information 130 may indicate a linguistic preference of the user 50. For example, the language context information 130 may indicate whether the user 50 likes to use relatively large words, phrases or sentences over relatively small words, phrases or sentences. In some implementations, the language context information 130 may indicate a vocabulary of the user 50. For example, the language context information 130 may indicate whether the user 30 tends to use medical terms, legal terms or engineering terms when speaking. In some implementations, the language context information 130 may indicate a verbosity (e.g., a wordiness) of speech uttered by the user 50.

In some implementations, the second electronic device 150 transmits (e.g., streams) the first text representation 120 and the language context information 130 to the first electronic device 100. Audible signal data corresponding to the audio signal 122 may also be transmitted to the first electronic device 100. In some implementations, the second electronic device 150 transmits video signal data (e.g., a video feed) to the first electronic device 100. The video signal data may include, for example, a two-dimensional (2D) representation of the user 20.

As represented in FIG. 1B, in some implementations, the first electronic device 100 receives the first text representation 120 and the language context information 130, and determines a second text representation 140. The second text representation 140 is determined based on the first text representation 120, the language context information 130 associated with the second electronic device 150, and language context information 160 that is associated with the first electronic device 100.

In some implementations, the language context information 160 indicates a language that is associated with the first electronic device 100. For example, in some implementations, the language context information 160 indicates a default language of the first electronic device 100. In some implementations, the language context information 160 indicates a language preference of the user 20 of the first electronic device 100. In some implementations, the first electronic device 100 determines the language preference of the user 20 based on a language of content that the user 20 has previously viewed. In some implementations, the first electronic device 100 determines the language preference of the user 20 based on a language that the user 20 speaks in. In some implementations, the language context information 160 indicates a dialect of the user 20.

In some implementations, the language context information 160 indicates a linguistic characteristic (e.g., a linguistic preference) of the user 20. For example, in some implementations, the language context information 160 indicates whether or not the user 20 prefers to read slangs (e.g., “what's up?” instead of “what are you up to?”). In some implementations, the language context information 160 indicates whether the user 20 prefers to read relatively long words, phrases or sentences over relatively short words, phrases or sentences. In some implementations, the language context information 160 indicates whether the user 20 has a particular vocabulary. For example, the language context information 160 may indicate whether the user 20 prefers to read text with medical terms, legal terms or engineering terms. In some implementations, the language context information 160 indicates whether or not the user 20 prefers to read verbose (e.g., wordy) text.

In some implementations, the language context information 160 indicates a location that is associated with the first electronic device 100 (e.g., a home location of the first electronic device 100). The first electronic device 100 may determine the language context information 160 based on information available to the first electronic device 100. For example, in some implementations, the first electronic device 100 determines the language context information 160 based on a device profile that is associated with the first electronic device 100. The device profile may associate the first electronic device 100 with a geographic region, such as a country where a language is predominantly spoken or where a language is spoken in a particular manner (e.g., with a particular linguistic style or using a particular vocabulary). In some implementations, the first electronic device 100 determines the language context information 160 based on a user profile that is associated with the user 20. For example, the user profile may associate the user 20 with a geographic region, such as a country with a predominant language or a predominant speaking style. In some implementations, the first electronic device 100 determines the language context information 160 based on GPS information, which may be mapped to a geographic region with a predominant language or a predominant linguistic style. The first electronic device 100 may determine the language context information 160 based on a network identifier, such as a cell identifier or an SSID, which may be associated with a geographic region with a predominant language or a predominant speaking style. In some implementations, the first electronic device 100 determines the language context information 160 by monitoring content that the user 20 views on the first electronic device 100. For example, the first electronic device 100 can identify a set of linguistic preferences of the user 20 by determining linguistic characteristics (e.g., formality, vocabulary, verbosity, etc.) of content that the user 20 has previously viewed via the first electronic device 100.

In some implementations, the second text representation 140 is a transcription of the audio signal 122, e.g., the first text representation 120 may be used as the second text representation 140. In some implementations, the first electronic device 100 determines to generate a translation of the first text representation 120 into a target language. The first electronic device 100 may determine to generate the translation of the first text representation 120 in response to a user input that enables a translation mode. In some implementations, the first electronic device 100 determines to generate the translation in response to determining that the first electronic device 100 and the second electronic device 150 are associated with different languages, e.g., the language context information 130 and the language context information 160 indicate different languages. In some implementations, the first electronic device 100 determines to generate the translation in response to determining that the user 20 and the user 50 are associated with different preferred languages. In some implementations, the first electronic device 100 determines to generate the translation in response to determining that the audio signal 122 or the first text representation 120 is in a different language than a preferred language of the user 20 or a default language of the first electronic device 100.

The target language may be associated with the language context information 160. In some implementations, the target language is associated with the user 20, independently of a location of the first electronic device 100. For example, the user 20 may specify that the user 20 wishes to read the second text representation 140 in a preferred language that is different from a language associated with a current location of the first electronic device 100.

In some implementations, the first electronic device 100 generates the second text representation 140 as a function of audible signal data that corresponds to the audio signal 122. For example, if the first text representation 120 is unavailable, unreliable, or is determined to contain inaccurate information, the first electronic device 100 may generate the second text representation 140 using the audible signal data to reconstruct the dialog. In some implementations, the first electronic device 100 generates an audio output signal based on the second text representation 140. For example, if the audible signal data corresponding to the audio signal 122 is of poor audio quality, the first electronic device 100 may use the second text representation 140 to synthesize a replacement audio signal to be output, e.g., from speakers associated with the first electronic device 100.

After generating the second text representation 140, the first electronic device 100 may cause the second text representation 140 to be displayed. In some implementations, the second text representation 140 is displayed in the XR environment 106. The second text representation 140 may be displayed in connection with (e.g., adjacent to) a representation of the user 50. In some implementations, the second text representation 140 is displayed in a designated area of the XR environment 106, e.g., at a lower area of the field of view. In some implementations, the first electronic device 100 causes an audio version of the second text representation 140 to be played via a speaker.

In some implementations, the first electronic device 100 includes or is attached to a head-mountable device (HMD) worn by the user 20. The HMD presents (e.g., displays) the XR environment 106 according to various implementations. In some implementations, the HMD includes an integrated display (e.g., a built-in display) that displays the XR environment 106. In some implementations, the HMD includes a head-mountable enclosure. In various implementations, the head-mountable enclosure includes an attachment region to which another device with a display can be attached. For example, in some implementations, the first electronic device 100 can be attached to the head-mountable enclosure. In various implementations, the head-mountable enclosure is shaped to form a receptacle for receiving another device that includes a display (e.g., the first electronic device 100). For example, in some implementations, the first electronic device 100 slides/snaps into or otherwise attaches to the head-mountable enclosure. In some implementations, the display of the device attached to the head-mountable enclosure presents (e.g., displays) the XR environment 106. In various implementations, examples of the first electronic device 100 include smartphones, tablets, media players, laptops, etc.

FIG. 2 illustrates a block diagram of the content presentation engine 200 in accordance with some implementations. In some implementations, the content presentation engine 200 includes a data obtainer 210, a text determiner 220, and a text compositor 230. In various implementations, the data obtainer 210 receives a first text representation 212 of audible signal data (hereafter “audio signal 214”) that corresponds to a spoken phrase detected at a second device (e.g., the first text representation 120 of the audio signal 122 shown in FIG. 1A). For example, a user of the second electronic device may participate in a communication session with a user of a device with which the content presentation engine 200 is associated. The communication session may include audio and/or video components. The second electronic device may include an audio sensor that detects the audio signal 214. In some implementations, the second electronic device generates the first text representation 212 based on the audio signal 214, for example, using speech-to-text conversion. The first text representation 212 may include cues separate from the spoken text of the communication session. For example, the first text representation 212 may include closed captions that include indicators of non-speech sounds and/or indicators of the identity of a speaker. In some implementations, the second electronic device generates the first text representation 212 in response to a user input that enables an audio visualization mode (e.g., a transcript mode or a closed caption mode). The user input may be received from one or more of the participants in the communication session.

In some implementations, the data obtainer 210 obtains (e.g., receives or determines) second language context information 216 associated with the second electronic device (e.g., the language context information 130 shown in FIGS. 1A and 1B). In some implementations, the data obtainer 210 obtains (e.g., receives or determines) first language context information 218 associated with the first electronic device (e.g., the language context information 160 shown in FIG. 1B).

In some implementations, the text determiner 220 determines a second text representation 222 (e.g., the second text representation 140 shown in FIG. 1B) based on the first text representation 212, the first language context information 218 associated with the device with which the content presentation engine 200 is associated, and the second language context information 216 that is associated with the second device. The second language context information 216 may indicate a language that is associated with the second device or a user of the second device. The second language context information 216 may indicate a location that is associated with the second device. The second language context information 216 may be determined by the second device or the data obtainer 210, for example, based on information available to the second device or the data obtainer 210, such as a device profile that is associated with the second device or a user profile that is associated with a user of the second device. The device profile or the user profile may associate the second device or the user of the second device with a particular language, a dialect, a set of linguistic preferences and/or a particular geographic region such as a country where a particular language is spoken. In some implementations, the second language context information 216 is determined based on GPS information, which may be mapped to a geographic region where a particular language is spoken. In some implementations, the second language context information 216 is determined based on a network identifier, such as a cell identifier or an SSID, which may be associated with a geographic region where a particular language is spoken.

In some implementations, the second language context information 216 is determined based on the audio signal 214. For example, the second device or the data obtainer 210 may detect a spoken language associated with a spoken phrase based on the audio signal 214. The second device or the data obtainer 210 may determine the second language context information 216 based on the spoken language. In some implementations, the second device or the data obtainer 210 determines the second language context information 216 based on the first text representation 212. For example, the second device or the data obtainer 210 may analyze the first text representation 212 to determine a language represented in the first text representation 212. The second device or the data obtainer 210 may determine the second language context information 216 based on the language. In some implementations, the second language context information 216 can be determined by the first device in the same way as described above.

The second language context information 216 may include geographic information, such as a country identifier and/or an identifier of a political subdivision of a country, such as a province or a state where a particular language is spoken. For example, the second language context information 216 may specify that the second device is associated with a majority-English or a majority-French speaking province of Canada. In some implementations, the second language context information 216 additionally or alternatively includes a language identifier. For example, the second language context information 216 may specify that the user resides in Canada and speaks English or French.

In some implementations, the data obtainer 210 obtains the second language context information 216 from the second device. In some implementations, the data obtainer 210 obtains the second language context information 216 from a database (e.g., the database 132 shown in FIGS. 1A and 1B). The database may store data associating electronic devices and/or users with respective geographic regions where respective languages are spoken. In some implementations, the data obtainer 210 determines (e.g., generates) the second language context information 216 in the same way as described above.

In some implementations, the data obtainer 210 obtains (e.g., receives) the first text representation 212 and the second language context information 216 from the second device. The data obtainer 210 may also obtain the audio signal 214 from the second device. In some implementations, the data obtainer 210 obtains (e.g., receives) a video signal data from the second device. The video signal data may include, for example, a representation of a user of the second device.

In some implementations, the text determiner 220 determines a second text representation 222 corresponding to the first text representation 212 or audio signal 214. The second text representation 222 may be determined based on the first text representation 212, first language context information 218 associated with the content presentation engine 200, and the second language context information 216.

In some implementations, the first language context information 218 indicates a language that is associated with the first device (e.g., the first electronic device 100 shown in FIGS. 1A and 1B). For example, as described in relation to FIG. 1B, in some implementations, the first language context information 218 may indicate a default language of the first device or a preferred language of a user of the first device. In some implementations, the first language context information 218 indicates a linguistic style associated with the first device (e.g., linguistic preference(s) of the user of the first device). For example, the first language context information 218 may indicate whether the user likes to read slangs (e.g., “what's up?”) or formal phrases (e.g., “what are you up to?”). As another example, the first language context information 218 may indicate whether the user prefers to read full words or abbreviations. As another example, the first language context information 218 may indicate a favorite book genre of the user (e.g., science fiction). In some implementations, the first language context information 218 indicates a language that is typically spoken in a home location of the first device. In some implementations, the content presentation engine 200 (e.g., the data obtainer 210) determines the first language context information 218 based on content that has been previously displayed on the display 232 (e.g., based on languages and/or linguistic characteristics of books, magazines or websites that the user has previously viewed). In some implementations, the content presentation engine 200 determines the first language context information 218 based on the user's previous speech. For example, the content presentation engine 200 may determine that the user prefers full words over abbreviations if the user's prior speech includes full words instead of abbreviations (e.g., “corporation” instead of “Corp.”).

The first language context information 218 may be determined based on information available to the content presentation engine 200. For example, in some implementations, the first language context information 218 is determined based on a device profile that is associated with the device in which the content presentation engine 200 is implemented. The device profile may be associated with a geographic region, such as a country where a particular language is spoken or where a language is spoken in a particular manner (e.g., with a particular vocabulary or with a particular linguistic style). In some implementations, the first language context information 218 is determined based on a user profile that is associated with a user. For example, the user profile may associate the user with a geographic region, such as a country with a predominant language or a predominant speaking style. In some implementations, the first language context information 218 is determined based on GPS information, which may be mapped to a geographic region with a predominant language or a predominant linguistic characteristic. The first language context information 218 may be determined based on a network identifier, such as a cell identifier or an SSID, which may be associated with a geographic region with a predominant language or a predominant linguistic characteristic.

In some implementations, e.g., if the first language context information 218 and the second language context information 216 are within a degree of similarity to one another, the second text representation 222 is a transcription of the audio signal 214, e.g., the first text representation 212 may be used as the second text representation 222. In some implementations, e.g., if the first language context information 218 and the second language context information 216 are at least a threshold level of difference from one another, the text determiner 220 determines to generate a translation of the first text representation 212 into a target language. The text determiner 220 may determine to generate the translation of the first text representation 212 in response to a user input that enables a translation mode. In some implementations, the text determiner 220 determines to generate the translation in response to determining that the content presentation engine 200 and the second device are associated with different locales, e.g., the first language context information 218 and the second language context information 216 are associated with different geographic regions where different languages are spoken or the same language is spoken with different speaking styles. In some implementations, the text determiner 220 determines to generate the translation in response to determining that the content presentation engine 200 and the second device are associated with different languages, e.g., the first language context information 218 and the second language context information 216 are associated with geographic regions associated with different languages or the same language with different speaking styles. In some implementations, the text determiner 220 determines to generate the translation in response to determining that the respective users of the content presentation engine 200 and the second device are associated with different locales, e.g., different geographic regions with different linguistic preferences. In some implementations, the text determiner 220 determines to generate the translation in response to determining that the first text representation 212 or the audio signal 214 corresponds to a different language than the target language. In some implementations, the first language context information 218 may specify a user-specified target language, for example, the user may specify that the user wishes to hear or read the second text representation 222 in a preferred language that is different from a language associated with a geographic region of the content presentation engine 200.

In some implementations, the second text representation 222 is in the same language as the first text representation 212 but may have spelling or vocabulary differences as compared with the first text representation 212. For example, if the first language context information 218 corresponds to the United States and the second language context information 216 corresponds to the United Kingdom, the text determiner 220 may replace United Kingdom spellings and vocabulary in the first text representation 212 with United States equivalents (e.g., replace “colour” with “color”, replace “traveller” with “traveler”, replace “flat” with “apartment”, etc.).

In some implementations, the text determiner 220 generates the second text representation 222 as a function of the audio signal 214. For example, if the first text representation 212 is unavailable, unreliable, or is determined to contain inaccurate information, the text determiner 220 may generate the second text representation 222 using the audio signal 214 to reconstruct the dialog. In some implementations, the text determiner 220 may use the second language context information 216 to perform speech-to-text conversion using a language model that is specific to a locale of the audio signal 214. In some implementations, the text determiner 220 generates an audio output signal based on the second text representation 222. For example, if the audio signal 214 is of poor audio quality, the text determiner 220 may use the second text representation 222 to synthesize a replacement audio signal to be output.

In some implementations, the first language context information 218 indicates a first dialect and the second language context information 216 indicates a second dialect that is different from the first dialect. The text determiner 220 generates the second text representation 222 by converting the first text representation 212 from the second dialect to the first dialect.

In some implementations, the first language context information 218 indicates a first linguistic characteristic and the second language context information 216 indicates a second linguistic characteristic that is different from the first linguistic characteristic. The text determiner 220 generates the second text representation 222 by changing a linguistic characteristic of the first text representation 212 from the second linguistic characteristic to the first linguistic characteristic. Examples of linguistic characteristics include vocabulary (e.g., U.S. English vs. U.K. English), literary style (e.g., modern English vs. Shakespearean English), formality (e.g., formal language vs. informal language with slangs) and verbosity (e.g., wordiness).

In some implementations, the text compositor 230 causes the second text representation 222 to be displayed on a display 232. The text compositor 230 may composite the second text representation 222 with video data 234 representing the video content of a communication session to generate video output data that is output to the display 232. In some implementations, the second text representation 222 is displayed in an XR environment. The second text representation 222 may be displayed in connection with a representation of a user, such as the user corresponding to the audio signal 214. In some implementations, the second text representation 222 is displayed in a designated area of the XR environment, e.g., at a lower area of the field of view. In some implementations, the second text representation 222 is overlaid adjacent to or onto a representation of a person (e.g., the user 50 shown in FIGS. 1A and 1B) that is uttering the spoken phrase. In some implementations, a portion of the second text representation 222 may be displayed as additional portions of the second text representation 222 are being generated by the text determiner 220. For example, the content presentation engine 200 may cause the display 232 to display a text representation that corresponds to a beginning portion of a sentence while the text determiner 220 is generating a text representation for a remainder of the sentence. Additionally, previously displayed portions of second text representation 222 may be modified or corrected by the text determiner 220 thereby generating an updated second text representation 222. For example, a determined text representation of the beginning of a sentence may change based on subsequently acquired knowledge of the later portions of the sentence.

In some implementations, the content presentation engine 200 may additionally or alternatively output an audio version of second text representation 222 via one or more speakers. For example, the content presentation engine 200 may perform a text-to-speech process on the second text representation 222 using a default or user-selected voice. The audio version may be output in mono, stereo, or spatialized such that it is perceived to emanate from a position in the environment (e.g., at a location corresponding to a user of the second device). In some implementations, a portion of second text representation 222 may be output as audio as additional portions of second text representation 222 are being generated by the text determiner 220. For example, the content presentation engine 200 may cause one or more speakers to output an audio version of a text representation that corresponds to a beginning portion of a sentence while the text determiner 220 is generating a text representation for a remainder of the sentence. In other implementations, the content presentation engine 200 may refrain from outputting an audio version of second text representation 222 until a completed version of the second text representation 222 is available. For example, the content presentation engine 200 may play an audio version of a sentence or a phrase after receiving the second text representation 222 for the entire sentence or the entire phrase.

FIGS. 3A-3B are a flowchart representation of a method 300 for presenting text representations based on language context in accordance with various implementations. In various implementations, the method 300 is performed by a device (e.g., the first electronic device 100 shown in FIGS. 1A-1B, or the content presentation engine 200 shown in FIGS. 1A-1B and 2). In some implementations, the method 300 is performed by processing logic, including hardware, firmware, software, or a combination thereof. In some implementations, the method 300 is performed by a processor executing code stored in a non-transitory computer-readable medium (e.g., a memory). In various implementations, an XR environment corresponding to a field of view of an image sensor (e.g., a scene-facing camera) of the device is displayed.

Briefly, the method 300 includes receiving a first text representation of an audio signal corresponding to a spoken phrase detected at a second device and determining a second text representation based on the first text representation, first language context information associated with the first device, and second language context information associated with the second device. The second text representation is presented via an output device (e.g., displayed on a display or an audio version of the text representation is played by a speaker).

In various implementations, as represented by block 310, the method 300 includes receiving a first text representation of audible signal data corresponding to a spoken phrase detected at a second device (e.g., receiving the first text representation 120 shown in FIGS. 1A and 1B, and/or the first text representation 212 shown in FIG. 2). For example, a user of the second device may participate in a local (e.g., collocated in the same physical environment) or remote (e.g., located in different physical environments) communication session that includes audio and/or video components. The second device may include an audio sensor that detects the audible signal and converts the audible signal to audible signal data. In some implementations, the second device generates the first text representation based on the audible signal data, for example, using speech-to-text conversion. The first text representation may include closed captions that includes cues separate from the spoken phrase of the communication session, such as indicators of non-speech sounds and/or indicators of the identity of a speaker. In some implementations, the second device generates the first text representation in response to a user input that enables an audio visualization mode (e.g., a transcript mode or a closed caption mode).

In various implementations, as represented by block 320, the method 300 includes determining a second text representation based on the first text representation, first language context information associated with a first device, and second language context information associated with the second device (e.g., determining the second text representation 140 shown in FIG. 1B and/or the second text representation 222 shown in FIG. 2). As represented by block 320a, the second language context information may be received from the second device. The second language context information may indicate a language that is associated with the second device (e.g., a default language of the second device and/or a preferred language of a user of the second device). The second language context information may indicate a linguistic characteristic associated with the second device (e.g., a formality level, a literary style, a vocabulary and/or a verbosity level). The second language context information may correspond to a location that is associated with the second device and may be determined by the second device, for example, based on information available to the second device. For example, as represented by block 320b, the second language context information may be determined based on a device profile that is associated with the second device. As represented by block 320c, in some implementations, the second language context information is determined based on a user profile that is associated with a user of the second device. The device profile or the user profile may associate the second device or the user of the second device with a language and/or a geographic region, such as a country with a predominant language or a predominant speaking style. In some implementations, as represented by block 320d, the second language context information is determined based on GPS information, which may be mapped to a geographic region where a particular language is spoken or a particular vocabulary is used. In some implementations, as represented by block 320e, the second language context information is determined based on a network identifier, such as a cell identifier or an SSID, which may be associated with a geographic region where a particular language is spoken or particular phrases are commonly used.

In some implementations, as represented by block 320f, the second language context information is determined based on the audible signal data. For example, the first device or the second device may detect a spoken language associated with a spoken phrase based on the audible signal data. The second device may determine the second language context information based on the spoken language. In some implementations, as represented by block 320g, the second device determines the second language context information based on the first text representation. For example, the second device may analyze the first text representation to determine a language represented in the first text representation. The second device may determine the second language context information based on the language.

In some implementations, as represented by block 320h, the second language context information is obtained from a database. The database may store data associating electronic devices and/or users with respective languages, linguistic characteristics and/or geographic regions with known predominant languages or speaking styles (e.g., commonly used slangs).

In some implementations, as represented by block 320i, determining the second text representation includes determining whether the first language context information and the second language context information satisfy a similarity threshold. If not, a translation of the first text representation is generated. In some implementations, as represented by block 320j, a target language for the translation is determined based on a language indicated by the first language context information. As represented by block 320k, in some implementations, a source language for the translation is determined based on a language indicated by the second language context information.

In some implementations, the first language context information indicates a language that corresponds to a location that is associated with a device in which the content presentation engine 200 is implemented. The first language context information may be determined based on information available to the content presentation engine 200. For example, in some implementations, as represented by block 320l, the first language context information is determined based on a device profile that is associated with the device in which the content presentation engine 200 is implemented. The device profile may be associated with a geographic region, such as a country where a particular language is spoken or a particular vocabulary is used. In some implementations, as represented by block 320m on FIG. 3B, the first language context information is determined based on a user profile that is associated with a user. For example, the user profile may associate the user with a geographic region, such as a country where a particular language is spoken or a particular vocabulary is used. In some implementations, as represented by block 320n, the first language context information is determined based on GPS information, which may be mapped to a geographic region where a particular language is spoken or a particular vocabulary is used. As represented by block 3200, in some implementations, the first language context information may be determined based on a network identifier, such as a cell identifier or an SSID, which may be associated with a geographic region where a particular language is spoken or a particular vocabular is used.

In some implementations, the second text representation is in the same language as the first text representation but may have spelling or vocabulary differences as compared with the first text representation. For example, if the first language context information corresponds to the United States and the second language context information corresponds to the United Kingdom, the device may replace United Kingdom spellings and vocabulary in the first text representation with United States equivalents. As another example, if the first text representation corresponds to speech spoken by a resident of the United Kingdom and the second text representation is being displayed on a device that belongs to an American, the device can replace “tube” with “subway” so that the American can more easily understand the reference to an underground train.

In some implementations, the first language context information indicates a first language associated with the first device and second language context information indicates a second language associated with the second device. In such implementations, determining the second text representation includes translating the first text representation from the second language to the first language. For example, translating the first text representation from a source language to a target language (e.g., a preferred language of a user of the first device).

In some implementations, the first language context information indicates a first locale associated with the first device and the second language context information indicates a second locale associated with the second device. In such implementations, determining the second text representation includes modifying the first text representation based on linguistic differences between the first locale and the second locale. For example, modifying the first text representation to generate the second text representation such that the second text representation appears to be a transcript of speech that is spoken in the first locale of the first device.

In some implementations, the first language context information indicates a first vocabulary associated with the first device and the second language context information indicates a second vocabulary associated with the second device. In such implementations, determining the second text representation includes replacing a phrase in the first text representation with a replacement phrase from the first vocabulary. For example, replacing phrases in the first text representation to generate the second text representation such that the second text representation uses phrases from the first vocabulary that the user of the first device may be more familiar with.

In some implementations, the first language context information indicates a first verbosity associated with the first device and the second language context information indicates a second verbosity associated with the second device. In such implementations, determining the second text representation includes modifying the first text representation to match the first verbosity associated with the first device. For example, adjusting a length of the first text representation to generate the second text representation such that the second text representation has a length (e.g., a word count) that the user of the first device prefers.

In some implementations, determining the second text representation includes determining whether the first language context information and the second language context information satisfy a similarity threshold. In some implementations, on a condition that the first language context information and the second language context information satisfies the similarity threshold, using the first text representation as the second text representation.

As represented by block 330, in various implementations, the method 300 includes presenting the second text representation (e.g., displaying the second text representation on a display, and/or generating and playing an audio version of the second text representation via a speaker). For example, as shown in FIG. 1B, the first electronic device 100 displays the second text representation 140 on a display of the first electronic device 100. In some implementations, the method 300 includes displaying a partially completed version of the text representation. For example, the device can display a second text representation that corresponds to a portion of a sentence while the device is waiting to receive the first text representation for a remainder of the sentence. By contrast, in some implementations, the method 300 includes outputting an audio version of a completed text representation. For example, the device plays an audio version of a sentence or a phrase after receiving the first text representation for the entire sentence or the entire phrase.

As represented by block 330a, in some implementations, the method 300 includes generating an audio representation of the second text representation. As represented by block 330b, in some implementations, the method 300 includes presenting the audio representation of the second text representation via an audio output device. For example, the first electronic device 100 can perform a text-to-speech operation to generate an audible version of the second text representation 140 and output the audible version via a speaker of the device. In some implementations, the device may generate the audio representation in response to an indication that a user of the device is visually-impaired and may not be able to properly view the second text representation. In some implementations, the device generates and outputs the audio representation of the second text representation when the device is in an accessibility mode. In some implementations, the device generates the audio representation when the audible signal data is distorted (e.g., unintelligible).

In some implementations, as represented by block 340, an audio input is received at the first device. A text representation of the audio input may be generated, as represented by block 340a. For example, the first device may use speech-to-text conversion to generate the text representation. As represented by block 340b, in some implementations, the audio input, the text representation of the audio input, and the first language context information are sent to the second device. A second text representation of the audio input may be generated based on the second language context information, as represented by block 340c. In some implementations, as represented by block 340d, the second text representation is sent to the second device. As represented by block 340e, the second text representation may be accompanied by an audio output signal. For example, the second text representation may be accompanied by synthesized speech. In some implementations, as represented by block 340f, the second text representation includes a text translation of the first text representation.

In some implementations, the method 300 includes receiving, from the second device, device status data that indicates a resource availability of the second device. In such implementations, the first device can generate the text representation of the audio input in response to the resource availability of the second device being below a resource availability threshold associated with generating the text representation. Generating the text representation may require a threshold amount of computing power, a threshold amount of memory and/or a threshold amount of battery level. If the device status data indicates that the second device does not have the required amount of computing power, memory and/or battery level, the first device determines that the second device may not have the capability to generate the text representation of the audio input. As such, the first device generates the text representation and transmits the text representation in addition to or instead of transmitting audible signal data that corresponds to the audio input.

In some implementations, the method 300 includes reconstructing audible signal data based on the text representation. In some implementations, the device receives audible signal data that corresponds to a phrase spoken by a user of another device. The device also receives a text representation (e.g., a transcript) of the phrase spoken by the user of the other device. In some implementations, the audible signal data may be distorted beyond a distortion threshold (e.g., corrupted, for example, due to errors introduced during transmission). In such implementations, the device can restore (e.g., correct the errors) the audible signal data based on the text representation. Since the device has a transcript of what was spoken at the other device, the device can use the transcript to reconstruct the distorted audio and play a reconstructed version of the audible signal data. In some implementations, the device generates new audible signal data based on the transcript and outputs the new audible signal data via a speaker of the device. In some implementations, the new audible signal data is a reconstruction of the audible signal data that the device received. The device can display a notification indicating that the audio being played has been altered (e.g., reconstructed) based on the transcript.

In some implementations, the method 300 includes reconstructing image data based on the text representation. In some implementations, the device receives image data (e.g., a video feed with a sequence of image frames) that depicts movement of a person that is speaking phrases that correspond to the text representation. The device also receives a text representation (e.g., a transcript) of the phrases spoken by the person. In some implementations, the image data may be distorted beyond a distortion threshold (e.g., corrupted, for example, due to errors introduced during transmission and/or due to limited bandwidth availability). In such implementations, the device can restore the image data based on the text representation. Since the device has a transcript of what was spoken at the other device, the device can use the transcript to reconstruct (e.g., correct the errors) the distorted image data and display a reconstructed version of the image data. In some implementations, the device generates new image data based on the transcript and displays the new image data via the display of the device. In some implementations, the new image data is a reconstruction of the image data that the device received. As an example, if the device receives a video feed of a person talking and a transcript of what the person is saying, the device can use the transcript to reconstruct the video feed if the video feed freezes. For example, if the transcript indicates that the person is talking and the video feed shows no movement in the person's lips, the device can simulate movement of the lips based on the transcript in order to provide an appearance that the lips are moving to utter the phrases included in the transcript. The device can display a notification indicating that the video feed being displayed has been altered (e.g., reconstructed) based on the transcript.