Maintaining original volume changes of a character in revoiced media stream
Methods, systems, and computer-readable media for artificially generating a revoiced media stream and maintaining original volume changes of a character in the revoiced media stream are provided. For example, a media stream including an individual speaking may be obtained. A transcript of the media stream may be obtained. The transcript of the media stream may be translated to a target language. A revoiced media stream in which the translated transcript in the target language is spoken by a virtual entity may be generated, wherein a ratio of the volume levels between first and second sets of words in the revoiced media stream is substantially identical to the ratio of volume levels between corresponding first and second utterances in the received media stream.
1. A computer program product for artificially generating a revoiced media stream, the computer program product embodied in a non-transitory computer-readable medium and including instructions for causing at least one processor to execute a method comprising:
receiving a media stream including a first individual and a second individual speaking in an origin language;
obtaining a transcript of the media stream including a first utterance and a second utterance spoke in the original language;
translating the transcript of the media stream to a target language, wherein the translated transcript includes a first set of words in the target language that corresponds with the first utterance and a second set of words in the target language that corresponds with the second utterance;
analyzing the media stream to determine at least one voice profile, wherein the at least one voice profile is indicative of a ratio of volume levels between the first utterance as spoken by the first individual and the second utterance as spoken by the second individual in the media stream;
determining metadata information for the translated transcript, wherein the metadata information includes desired volume levels for each of the first and second sets of words that correspond with the first and second utterances; and
using the determined at least one voice profile, the translated transcript, and the metadata information to artificially generate a revoiced media stream in which the first individual and the second individual sound as they speak the translated transcript, wherein a ratio of the volume levels between the first and second sets of words in the revoiced media stream is substantially identical to the ratio of volume levels between the first and second utterances in the received media stream.
2. The computer program product of claim 1 , wherein the at least one voice profile is further indicative of intonation differences between the first utterance as spoken by the first individual and the second utterance as spoken by the second individual in the media stream and the method further comprising:
generating a revoiced media stream in which intonation differences between the first and second sets of words in the revoiced media stream are substantially identical to the intonation differences between the first and second utterances in the received media stream.
3. The computer program product of claim 1 , wherein the method is further comprising:
determining that the first utterance as spoken by the first individual was pronounced as a question and that the second utterance as spoken by the second individual was pronounced as a statement; and
generating a revoiced media stream in which the first set of words are pronounced as a question and the second set of words are pronounced as a statement.
4. The computer program product of claim 1 , wherein the at least one voice profile is further indicative of pitch differences between the first utterance as spoken by the first individual and the second utterance as spoken by the second individual in the media stream and the method further comprising:
generating a revoiced media stream in which pitch differences between the first and second sets of words in the revoiced media stream are substantially identical to the pitch differences between the first and second utterances in the received media stream.
5. The computer program product of claim 1 , wherein the at least one voice profile is further indicative of accent differences between the first utterance as spoken by the first individual and the second utterance as spoken by the second individual in the media stream and the method further comprising:
generating a revoiced media stream in which accent differences between the first and second sets of words in the revoiced media stream are substantially identical to the accent differences between the first and second utterances in the received media stream.
6. The computer program product of claim 1 , wherein the method is further comprising:
determining that the first individual shouted the first utterance and that the second individual whispered the second utterance; and
generating a revoiced media stream that sounds as the first individual shouts the first set of words in the target language and the second individual whispers the second set of words in the target language.
7. The computer program product of claim 1 , wherein the method is further comprising:
determining that that the first individual spoke the first utterance in a cynical voice and that the second individual spoke the second utterance in a regular voice; and
generating a revoiced media stream that sounds as the first individual pronounces the first set of words in the target language in a cynical voice and the second individual pronounces the second set of words in the target language in a regular voice.
8. The computer program product of claim 1 , wherein the method is further comprising:
analyzing the media stream to determine volume levels for the first utterance as spoken by the first individual and the second utterance as spoken by the second individual in the media stream; and
generating a revoiced media stream in which the first and second sets of words are spoken in the target language at the determined levels of volume.
9. The computer program product of claim 1 , wherein the method is further comprising:
analyzing the media stream to determine volume levels for the first utterance as spoken by the first individual and the second utterance as spoken by the second individual in the media stream; and
generating a revoiced media stream in which the first and second sets of words are spoken in the target language at lower levels of volume than the determined level of volume in the media stream.
10. The computer program product of claim 1 , wherein the method is further comprising:
analyzing the media stream to determine volume levels for the first utterance as spoken by the first individual and the second utterance as spoken by the second individual in the media stream; and
generating a revoiced media stream in which the first and second sets of words are pronounced in the target language at higher levels of volume than the determined level of volume in the media stream.
11. The computer program product of claim 1 , wherein the method is further comprising:
accessing user settings defining minimum and maximum volume levels in a revoiced media stream; and
determining to deviate from the ratio of volume levels between the first utterance as spoken by the first individual and the second utterance as spoken by the second individual in the received media stream based on user settings.
12. The computer program product of claim 1 , wherein the method is further comprising:
accessing language settings associated with the target language; and
determining to deviate from the ratio of volume levels between the first utterance as spoken by the first individual and the second utterance as spoken by the second individual in the received media stream based on language settings.
13. The computer program product of claim 1 , wherein the received media stream is a real-time conversation between the first individual, the second individual, and a particular user associated with the target language, and the method further comprising:
maintaining the ratio of the volume levels between the first and second sets of words in the revoiced media stream substantially identical to the ratio of volume levels between the first and second utterances in the real-time conversation.
14. The computer program product of claim 13 , wherein the method is further comprising:
recognizing in real-time that the first utterance has no real meaning in the original language and determining to mute the first utterance.
15. A system for artificially generating a revoiced media stream, the system comprising:
at least one processor configured to:
receive a media stream including a first individual and a second individual speaking in an origin language;
obtain a transcript of the media stream including a first utterance and a second utterance spoke in the original language;
translate the transcript of the media stream to a target language, wherein the translated transcript includes a first set of words in the target language that corresponds with the first utterance and a second set of words in the target language that corresponds with the second utterance;
analyze the media stream to determine at least one voice profile, wherein the at least one voice profile is indicative of a ratio of volume levels between the first utterance as spoken by the first individual and the second utterance as spoken by the second individual in the media stream;
determine metadata information for the translated transcript, wherein the metadata information includes desired volume levels for each of the first and second sets of words that correspond with the first and second utterances; and
use the determined at least one voice profile, the translated transcript, and the metadata information to artificially generate a revoiced media stream in which the first individual and the second individual sound as they speak the translated transcript, wherein a ratio of the volume levels between the first and second sets of words in the revoiced media stream is substantially identical to the ratio of volume levels between the first and second utterances in the received media stream.
16. The system of claim 15 , wherein the at least one processor is further configured to:
determine a first synthesized voice for a first virtual entity intended to dub the first individual, wherein the first synthesized voice has characteristics identical to the characteristics of a particular voice of the first individual;
determine a second synthesized voice for a second virtual entity intended to dub the second individual, wherein the second synthesized voice has characteristics identical to the characteristics of a particular voice of the second individual; and
generate a revoiced media stream in which the translated transcript in the target language is spoken by the first and second virtual entities.
17. The system of claim 15 , wherein the at least one the voice profile is indicative of intonation differences between the first utterance as spoken by the first individual and the second utterance as spoken by the second individual, and the at least one processor is further configured to:
generate a revoiced media stream in which intonation differences between the first and second sets of words in the revoiced media stream is substantially identical to the intonation differences between the first and second utterances as recorded the received media stream.
18. The system of claim 17 , wherein the at least one processor is further configured to:
determine that the first utterance was pronounced by the first individual as a question and the second utterance was pronounced by the second individual as an answer; and
generate a revoiced media stream in which the first set of words in the target language are pronounced as a question and the second set of words in the target language are pronounced as an answer.