Abstract: An audio processing method includes: acquiring an audio signal including audio frames; inputting the audio frames into a streaming acoustic network to obtain phoneme features representing phoneme information of the audio signal and streaming audio features; acquiring an entity set including first entities, wherein the first entities correspond to pieces of phoneme information; extracting second entities from the entity set based on the phoneme features, wherein the second entities correspond to the phoneme features, and wherein a second number of the second entities is greater than or equal to a third number of the audio frames and less than or equal to a first number of the first entities; obtaining a text recognition result based on inputting the audio signal, the streaming audio features, and the second entities into a non-streaming acoustic network; and outputting the text recognition result.

Abstract: An electronic device has a radiator including a first end, a second end, a ground point, a first connection point, and a second connection point. The ground point is disposed in a central area of the radiator. A length of the radiator is greater than three-quarters of a first wavelength which is a medium wavelength of a first resonance generated by the radiator. A first inductor is electrically connected between the first connection point and the ground plane, and a second inductor is electrically connected between the second connection point and the ground plane, inductance values of the inductors are less than or equal to a first threshold. A distance between the first connection point and the first end is less than a quarter of the first wavelength, and the second connection point is located between the first connection point and the second end.

Abstract: A speech recognition method includes: performing phoneme recognition on a speech signal to obtain a phoneme recognition result corresponding to a speech frame in the speech signal, the phoneme recognition result indicating a probability distribution of the corresponding speech frame in a phoneme space, and the phoneme space comprising a plurality of phonemes and a blank output; performing suppression adjustment on a probability of the blank output in the phoneme recognition result corresponding to the speech frame, to reduce a ratio of the probability of the blank output to a probability of a phoneme in the phoneme recognition result; and inputting the adjusted phoneme recognition result corresponding to the speech frame into a decoding map to obtain a recognition text sequence corresponding to the speech signal, the decoding map comprising a mapping relationship between characters and phonemes.

Abstract: An electronic device has a radiator including a first end, a second end, a ground point, a first connection point, and a second connection point. The ground point is disposed in a central area of the radiator. A length of the radiator is greater than three-quarters of a first wavelength which is a medium wavelength of a first resonance generated by the radiator. A first inductor is electrically connected between the first connection point and the ground plane, and a second inductor is electrically connected between the second connection point and the ground plane, inductance values of the inductors are less than or equal to a first threshold. A distance between the first connection point and the first end is less than a quarter of the first wavelength, and the second connection point is located between the first connection point and the second end.

Abstract: An audio processing method includes: acquiring an audio signal including audio frames; inputting the audio frames into a streaming acoustic network to obtain phoneme features representing phoneme information of the audio signal and streaming audio features; acquiring an entity set including first entities, wherein the first entities correspond to pieces of phoneme information; extracting second entities from the entity set based on the phoneme features, wherein the second entities correspond to the phoneme features, and wherein a second number of the second entities is greater than or equal to a third number of the audio frames and less than or equal to a first number of the first entities; obtaining a text recognition result based on inputting the audio signal, the streaming audio features, and the second entities into a non-streaming acoustic network; and outputting the text recognition result.

Abstract: A speech recognition method includes: performing phoneme recognition on a speech signal to obtain a phoneme recognition result corresponding to a speech frame in the speech signal, the phoneme recognition result indicating a probability distribution of the corresponding speech frame in a phoneme space, and the phoneme space comprising a plurality of phonemes and a blank output; performing suppression adjustment on a probability of the blank output in the phoneme recognition result corresponding to the speech frame, to reduce a ratio of the probability of the blank output to a probability of a phoneme in the phoneme recognition result; and inputting the adjusted phoneme recognition result corresponding to the speech frame into a decoding map to obtain a recognition text sequence corresponding to the speech signal, the decoding map comprising a mapping relationship between characters and phonemes.