Japanese CTC Model Configuration (CTC/CtcJaModels.swift)

PhoneClaw · Packages/FluidAudio/Sources/FluidAudio/ASR/Parakeet/SlidingWindow/CTC/CtcJaModels.swift · swift · Apache-2.0
ai, voice, lang:swift, tag:coreml, tag:japanese

Provides configuration settings for Japanese ASR models using CoreML. Use this to integrate speech-to-text capabilities into Swift applications targeting Japanese language processing.

// Source: https://github.com/kellyvv/PhoneClaw | License: Apache-2.0 | Copyright You must retain, in the Source form of any Derivative Works
// Provided via justclaw.it.com; keep this notice with the code.
@preconcurrency import CoreML
import Foundation

/// Configuration for Japanese CTC models
public enum CtcJaConfig: ParakeetLanguageModelConfig {
    public static let blankId: Int = 3072
    public static let repository: Repo = .parakeetCtcJa
    public static let languageLabel: String = "CTC ja (Japanese)"
    public static let loggerCategory: String = "CtcJaModels"

    public static let preprocessorFile: String = ModelNames.CTCJa.preprocessorFile
    public static let encoderFile: String = ModelNames.CTCJa.encoderFile
    public static let decoderFile: String = ModelNames.CTCJa.deco

... (truncated -- full source via MCP)

See the full source, get the GitHub permalink, and search for more like it.

Get a free API key