This repository has been archived on 2025-08-26. You can view files and clone it, but cannot push or open issues or pull requests.
Files
enginex_bi_series-sherpa-onnx/swift-api-examples/zipformer-ctc-asr.swift
Fangjun Kuang 3bf986d08d Support non-streaming zipformer CTC ASR models (#2340)
This PR adds support for non-streaming Zipformer CTC ASR models across 
multiple language bindings, WebAssembly, examples, and CI workflows.

- Introduces a new OfflineZipformerCtcModelConfig in C/C++, Python, Swift, Java, Kotlin, Go, Dart, Pascal, and C# APIs
- Updates initialization, freeing, and recognition logic to include Zipformer CTC in WASM and Node.js
- Adds example scripts and CI steps for downloading, building, and running Zipformer CTC models

Model doc is available at
https://k2-fsa.github.io/sherpa/onnx/pretrained_models/offline-ctc/icefall/zipformer.html
2025-07-04 15:57:07 +08:00

67 lines
1.7 KiB
Swift

import AVFoundation
extension AudioBuffer {
func array() -> [Float] {
return Array(UnsafeBufferPointer(self))
}
}
extension AVAudioPCMBuffer {
func array() -> [Float] {
return self.audioBufferList.pointee.mBuffers.array()
}
}
func run() {
let model = "./sherpa-onnx-zipformer-ctc-zh-int8-2025-07-03/model.int8.onnx"
let tokens = "./sherpa-onnx-zipformer-ctc-zh-int8-2025-07-03/tokens.txt"
let zipformerCtc = sherpaOnnxOfflineZipformerCtcModelConfig(
model: model
)
let modelConfig = sherpaOnnxOfflineModelConfig(
tokens: tokens,
debug: 0,
zipformerCtc: zipformerCtc
)
let featConfig = sherpaOnnxFeatureConfig(
sampleRate: 16000,
featureDim: 80
)
var config = sherpaOnnxOfflineRecognizerConfig(
featConfig: featConfig,
modelConfig: modelConfig
)
let recognizer = SherpaOnnxOfflineRecognizer(config: &config)
let filePath = "./sherpa-onnx-zipformer-ctc-zh-int8-2025-07-03/test_wavs/0.wav"
let fileURL: NSURL = NSURL(fileURLWithPath: filePath)
let audioFile = try! AVAudioFile(forReading: fileURL as URL)
let audioFormat = audioFile.processingFormat
assert(audioFormat.channelCount == 1)
assert(audioFormat.commonFormat == AVAudioCommonFormat.pcmFormatFloat32)
let audioFrameCount = UInt32(audioFile.length)
let audioFileBuffer = AVAudioPCMBuffer(pcmFormat: audioFormat, frameCapacity: audioFrameCount)
try! audioFile.read(into: audioFileBuffer!)
let array: [Float]! = audioFileBuffer?.array()
let result = recognizer.decode(samples: array, sampleRate: Int(audioFormat.sampleRate))
print("\nresult is:\n\(result.text)")
if result.timestamps.count != 0 {
print("\ntimestamps is:\n\(result.timestamps)")
}
}
@main
struct App {
static func main() {
run()
}
}