Abstract: Embodiments are provided for building a configurable multilingual model. A computing system obtains a plurality of language-specific automatic speech recognition modules and a universal automatic speech recognition module trained on a multi-language training dataset comprising training data corresponding to each of the plurality of different languages. The computing system then compiles the universal automatic speech recognition module with the plurality of language-specific automatic speech recognition modules to generate a configurable multilingual model that is configured to selectively and dynamically utilize a sub-set of the plurality of language-specific automatic speech recognition modules with the universal automatic speech recognition module to process audio content in response to user input identifying one or more target languages associated with the audio content.

Abstract: Disclosed are a method for recovering lithium from lithium iron phosphate waste and application thereof. The method comprises the following steps: (1) adding water to lithium iron phosphate waste to obtain a lithium iron phosphate slurry (2) adding a soluble iron salt to the lithium iron phosphate slurry to perform a reaction, and filtering a resulting product to obtain an iron phosphate slag and a filtrate containing Li+ and Fe2+; (3) adding an oxidizing agent to the filtrate to perform a reaction and filtering a resulting product to obtain iron hydroxide and a filtrate containing Li+, Fe3+; (4) performing a multi-stage counter-current circulation leaching to the mixture of the filtrate and the lithium iron phosphate waste to obtain a lithium solution. The present disclosure adopts a soluble iron salt capable of accelerating the conversion of lithium iron phosphate.

Abstract: Disclosed are a method for removing elemental copper from ternary battery waste and its application. The method comprises the following steps: crushing and screening the ternary battery waste to obtain a powder, and then removing iron by magnetic separation to obtain an iron-removed ternary waste; Adding an alkaline solution to the iron-removed ternary waste to perform an aluminum removal reaction, filtering to obtain a filter slag and aluminum-containing wastewater, washing the filter slag with water and drying to obtain a copper-nickel-cobalt-manganese material. Adding an iron salt solution to the copper-nickel-containing material to perform a leaching process, filtering to obtain a leachate and a nickel-cobalt-manganese waste; adding iron powder to the leachate and stirring to perform a reaction, filtering to obtain a copper residue, washing the copper residue with water and drying to obtain a copper-removed liquid and a sponge copper.

Abstract: This disclosure discloses a method for recovering lithium from a waste LFP material, including: S1. adding water to the waste LFP material, controlling a pH thereof at 0.5 to 2.0 and an ORP of the slurry at 0.05 V to 1.2 V, and filtering to obtain a material A; S2. adding sulfuric acid and heating a resulting mixture at 100° C. to 400° C. in the air or an oxygen atmosphere to obtain a material B; S3. adding water to the material B, and stirring and filtering to obtain a material C; S4. controlling a pH of the material C at 9 to 11, and filtering a resulting mixture to obtain a material D; S5. passing the material D through an ion-exchange resin to obtain a material E; and S6. adding the material E to a sodium carbonate solution to react; and collecting a resulting solid to obtain lithium carbonate.

Abstract: A computer device is provided that includes one or more processors configured to receive an end-to-end (E2E) model that has been trained for automatic speech recognition with training data from a source-domain, and receive an external language model that has been trained with training data from a target-domain. The one or more processors are configured to perform an inference of the probability of an output token sequence given a sequence of input speech features. Performing the inference includes computing an E2E model score, computing an external language model score, and computing an estimated internal language model score for the E2E model. The estimated internal language model score is computed by removing a contribution of an intrinsic acoustic model. The processor is further configured to compute an integrated score based at least on E2E model score, the external language model score, and the estimated internal language model score.

Abstract: A computer device is provided that includes one or more processors configured to receive an end-to-end (E2E) model that has been trained for automatic speech recognition with training data from a source-domain, and receive an external language model that has been trained with training data from a target-domain. The one or more processors are configured to perform an inference of the probability of an output token sequence given a sequence of input speech features. Performing the inference includes computing an E2E model score, computing an external language model score, and computing an estimated internal language model score for the E2E model. The estimated internal language model score is computed by removing a contribution of an intrinsic acoustic model. The processor is further configured to compute an integrated score based at least on E2E model score, the external language model score, and the estimated internal language model score.

Abstract: Computer-implemented systems and methods are provided for assessing non-native speech proficiency. A non-native speech sample is processed to identify a plurality of vowel sound boundaries in the non-native speech sample. Portions of the non-native speech sample are analyzed within the vowel sound boundaries to extract vowel characteristics. The vowel characteristics are used to identify a plurality of vowel space metrics for the non-native speech sample, and the vowel space metrics are used to determine a non-native speech proficiency score for the non-native speech sample.

Abstract: Computer-implemented systems and methods are provided for assessing non-native speech proficiency. A non-native speech sample is processed to identify a plurality of vowel sound boundaries in the non-native speech sample. Portions of the non-native speech sample are analyzed within the vowel sound boundaries to extract vowel characteristics. The vowel characteristics are used to identify a plurality of vowel space metrics for the non-native speech sample, and the vowel space metrics are used to determine a non-native speech proficiency score for the non-native speech sample.