Device and method for realizing simultaneous translation function of telephone

Abstract

The invention provides a device and method for realizing a simultaneous translation function of a telephone, and the device comprises a state monitoring module which is used for receiving a simultaneous translation request sent by a user based on a telephone terminal, and capturing the user voice signals of the telephone terminals of two communication parties in real time based on a receiving result; the simultaneous translation module is used for converting the user voice signal into a target text, performing target language real-time translation on the target text based on a translation demand, and converting a target language real-time translation result into a corresponding target voice; and the communication module is used for transmitting the target voice to the corresponding telephone terminals for output based on the required call direction of the target voice. The communication efficiency is improved, seamless communication between different language users is realized, and the simultaneous translation effect is guaranteed.

Claims

1. A device for realizing the simultaneous interpretation function of a telephone, characterized in that: include: A status monitoring module, used for receiving a simultaneous interpretation request sent by a user based on a telephone terminal, and capturing user voice signals of the telephone terminals of both parties of the call in real time based on the receiving result; The simultaneous interpretation module is used to convert the user's voice signal into a target text, and translate the target text into a target language in real time based on the translation requirements, and convert the target language real-time translation result into the corresponding target voice; The communication module is used to transmit the target voice to the corresponding telephone terminal for output based on the required call direction of the target voice.

2. The device for realizing simultaneous interpretation of telephone according to claim 1, characterized in that: The condition monitoring module includes: Simultaneous interpretation box, telephone handset and telephone host; The simultaneous interpretation box is electrically connected to the telephone handle and the telephone host via an RJ9 connection cable.

3. The device for realizing simultaneous interpretation of telephone according to claim 1, characterized in that: Condition monitoring module, including: A request monitoring unit, used to monitor a trigger signal of a user based on a preset function key on a telephone terminal, and determine a level change amplitude of the trigger signal, and when the level change amplitude meets a preset condition, determine that a simultaneous interpretation request sent by the user is received; The voice signal acquisition unit is used to control the transmitter and receiver on the telephone terminal based on the reception result to respectively collect the original voice signal of the local user and the voice signal of the remote user received from the telephone line and originally sent to the receiver for broadcast after preliminary internal processing, and input the collected original voice signal of the local user and the voice signal of the remote user into the simultaneous interpretation box respectively.

4. The device for realizing simultaneous interpretation of telephone according to claim 3, characterized in that: The voice signal acquisition unit comprises: The format conversion subunit is used to obtain the original voice signal of the local user collected by the microphone, and convert the original voice signal into a digital signal based on a preset analog-to-digital conversion circuit; The signal sending subunit is used to cross-input the digital signal corresponding to the original voice signal and the voice signal of the remote user into the simultaneous interpretation box based on the time sequence of the original voice signal of the local user and the voice signal of the remote user.

5. The device for realizing simultaneous interpretation of telephone according to claim 4, characterized in that: Format conversion subunit, including: A signal sampling subunit, used for sampling the original speech signal according to a fixed time interval based on a sampler to obtain discrete sample points; The conversion subunit is used to quantize the sampling values ​​of discrete sample points based on the quantization standard to obtain quantized sample values, and to represent the quantized sample values ​​based on binary codes to obtain digital signals corresponding to the original speech signals.

6. The device for realizing simultaneous interpretation of telephone according to claim 1, characterized in that: Simultaneous interpretation module, including: Signal processing unit for: Acquire the user voice signal, and perform denoising on the user voice signal; Extracting signal components of the user voice signal at different times based on the denoising result, determining average amplitude differences of the user voice signals in adjacent short time periods based on the signal components, and determining the starting endpoint and the ending endpoint of the user voice signal based on the average amplitude differences; Segmenting the user voice signal according to the start endpoint and the end endpoint based on the preset number of frames to obtain sub-voice signal segments, and determining the fundamental frequency information and spectrum characteristics of the sub-voice signal segments respectively, and obtaining the acoustic characteristics of the sub-voice signal segments based on the fundamental frequency information and the spectrum characteristics; Language and text conversion unit for: Matching the acoustic features with a preset language reference table, and determining the communication language corresponding to the user voice signal based on the match; The target language model is selected from the model library using the spoken language as a constraint condition, and each sub-speech signal segment is converted into text based on the target language model to obtain a corresponding text data set; Text translation units, for: Extracting the grammatical structure and vocabulary set of each text data in the text data set respectively, and determining the target semantics of the text data based on the grammatical structure and vocabulary set; Determine the target language translation direction of the text data based on the call requirements, determine the available corpus based on the target language translation direction, and query the translation text corresponding to each word in the available corpus based on the target semantics; Determine the grammatical rules of the corresponding language based on the translation direction, associate the word order of the translation text based on the grammatical rules, and obtain the complete translation text based on the word order association result; The simultaneous interpretation unit is used to convert the complete translation text into speech by adopting a corresponding translation mechanism based on the translation direction to obtain the corresponding target speech.

7. The device for realizing simultaneous interpretation of telephone according to claim 6, characterized in that: Simultaneous interpretation unit, including: A text acquisition subunit, used for acquiring the obtained complete translation text and dividing the complete translation text into morpheme sequences based on the grammatical rules of the complete translation text; The speech conversion subunit is used to: Determine the language expression feature of the user's voice signal based on the acoustic feature of the user's voice signal, and obtain a prosody reference index corresponding to the voice conversion based on the language expression feature; At the same time, the phoneme dictionary is accessed based on the morpheme sequence, and the phonemes corresponding to each morpheme in the morpheme sequence are obtained based on the access result; Based on the morpheme sequence, the phonemes corresponding to each morpheme are associated to obtain the corresponding phoneme sequence, and the phoneme sequence is adapted to the speech features based on the prosodic reference index; Based on the speech feature adaptation results, the complete translation text is subjected to electronic speech synthesis to obtain the corresponding target speech.

8. The device for realizing simultaneous interpretation of telephone according to claim 1, characterized in that: Communication module, including: A direction determination unit, used to obtain a source voice signal input end corresponding to the target voice, and obtain a required call direction of the target voice based on the source voice signal input end; The voice output unit is used to perform different signal output processing on the target voice using a difference processing mechanism based on the required call direction, and transmit the target voice to the corresponding telephone terminal for output based on the signal output processing result.

9. The device for realizing simultaneous interpretation of telephone according to claim 8, characterized in that: Voice output unit, comprising: The speech output processing subunit is used to: When the call direction is from a local user to a remote user: Modulate the target voice, and output the modulated target voice to the original telephone handset transmitter line, and transmit the target voice to the remote user's telephone terminal for output based on the output result; When the call direction is from the remote user to the local user: Obtaining the audio specifications of the original telephone, and performing reverse signal restoration of the target voice based on the audio specifications of the original telephone; The target voice after the reverse signal is restored is output on the local user's telephone terminal.

10. A method for realizing the simultaneous interpretation function of a telephone, characterized in that: include: Step 1: receiving a simultaneous interpretation request sent by a user based on a telephone terminal, and capturing user voice signals of the telephone terminals of both parties in the call in real time based on the receiving result; Step 2: Convert the user voice signal into a target text, and translate the target text into the target language in real time based on the translation requirements, and convert the target language real-time translation result into the corresponding target voice; Step 3: Based on the required call direction of the target voice, the target voice is transmitted to the corresponding telephone terminal for output.

Images