mlboydaisuke commited on
Commit
48d9e31
·
verified ·
1 Parent(s): b1a94d7

gen-cards: regenerate Use-it block

Browse files
Files changed (1) hide show
  1. README.md +42 -0
README.md CHANGED
@@ -25,6 +25,48 @@ Gated **77/77 token-exact** end-to-end vs the HF `ParakeetForTDT` reference, and
25
  through the Swift **CoreAIKit** `KitParakeetModel`. blank 8192 · durations [0,1,2,3,4] · 16 kHz.
26
 
27
  <!-- gen-cards:use-it begin id=parakeet-tdt-0.6b-v3 (managed by scripts/gen-cards — edit cards.json / QuickStart.swift, not this block) -->
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
28
  <!-- gen-cards:use-it end -->
29
 
30
  ## Use (CoreAIKit)
 
25
  through the Swift **CoreAIKit** `KitParakeetModel`. blank 8192 · durations [0,1,2,3,4] · 16 kHz.
26
 
27
  <!-- gen-cards:use-it begin id=parakeet-tdt-0.6b-v3 (managed by scripts/gen-cards — edit cards.json / QuickStart.swift, not this block) -->
28
+ ![Parakeet-TDT 0.6B v3 demo](https://huggingface.co/mlboydaisuke/Parakeet-TDT-0.6B-CoreAI/resolve/main/demo.gif)
29
+ *Parakeet-TDT 0.6B on iPhone 17 Pro — the zoo's coreai-audio app, real speed.*
30
+
31
+ ## Use it
32
+
33
+ ▶️ **Run it (source)** — the [Transcribe runner](https://github.com/john-rocky/coreai-kit/tree/main/Examples/Transcribe)
34
+ (GUI + CLI, one app for every speech-to-text model in the catalog):
35
+
36
+ ```bash
37
+ git clone https://github.com/john-rocky/coreai-kit
38
+ open coreai-kit/Examples/Transcribe/Transcribe.xcodeproj
39
+ # → Run, then pick "Parakeet-TDT 0.6B v3" in the model picker
40
+
41
+ # agents / headless (macOS):
42
+ cd coreai-kit/Examples/Transcribe
43
+ swift run transcribe-cli --model parakeet-tdt-0.6b-v3 --audio sample.wav
44
+ ```
45
+
46
+ 💻 **Build with it** — complete; the glue is kit API, copy-paste runs:
47
+
48
+ ```swift
49
+ import CoreAIKit
50
+
51
+ let transcriber = try await KitTranscriber(catalog: "parakeet-tdt-0.6b-v3")
52
+ let samples = try AudioFile.pcm16kMono(url) // any wav/m4a/mp3 → 16 kHz mono Float
53
+ let result = try await transcriber.transcribe(samples: samples)
54
+ // result.text (25 EU languages)
55
+ ```
56
+
57
+ The take-home is [`Examples/Transcribe/Sources/QuickStart.swift`](https://github.com/john-rocky/coreai-kit/blob/main/Examples/Transcribe/Sources/QuickStart.swift)
58
+ — this exact code as one typed function, no UI; both the runner's GUI and its CLI call it.
59
+ Recording? `MicRecorder` (kit API) captures mic audio as 16 kHz mono `[Float]` — the record
60
+ button and permission prompt are your app's own chrome.
61
+
62
+ **Integration checklist**
63
+
64
+ - SPM: `https://github.com/john-rocky/coreai-kit` → product **CoreAIKit**
65
+ - Info.plist: `NSMicrophoneUsageDescription` — only if you record
66
+ - Entitlements: none needed (macOS)
67
+ - First run downloads the model — 1.3 GB (Mac) — then it loads from the
68
+ local cache (Application Support; progress via the `downloadProgress` callback)
69
+ - Measure in Release — Debug is ~3× slower on per-token host work
70
  <!-- gen-cards:use-it end -->
71
 
72
  ## Use (CoreAIKit)