### Setup Development Environment

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/CONTRIBUTING.md

Use the Makefile to set up your development environment by installing necessary dependencies.

```bash
make setup
```

--------------------------------

### Setup Environment and Download Model

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Set up the necessary environment and download a specific model for use with the Swift CLI. Ensure git-lfs is installed before running.

```bash
make setup
make download-model MODEL=large-v3-v20240930_626MB
```

--------------------------------

### Install Python Client Dependencies

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Installs the necessary Python dependencies for the OpenAI SDK client example.

```bash
cd Examples/ServeCLIClient/Python
uv sync
```

--------------------------------

### Initialize and Run WhisperKit Client

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/Examples/ServeCLIClient/Python/README.md

Commands to start the server, install dependencies, and execute basic transcription or translation tasks.

```bash
whisperkit-cli serve
```

```bash
uv sync
```

```bash
# Transcribe an audio file
python whisperkit_client.py transcribe audio.wav

# Translate an audio file to English
python whisperkit_client.py translate audio.wav

# Test with sample files
python whisperkit_client.py test
```

--------------------------------

### Device Information Example

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/BENCHMARKS.md

Example output format for connected devices when running 'make list-devices'. Verify the 'state' is 'connected'.

```ruby
{ 
   :name=>"My Mac", 
   :type=>"Apple M2 Pro", 
   :platform=>"macOS", 
   :os_version=>"15.0.1", 
   :product=>"Mac14,12", 
   :id=>"XXXXXXXX-1234-5678-9012-XXXXXXXXXXXX", 
   :state=>"connected"
}
```

--------------------------------

### WhisperKit Client Usage Examples

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/Examples/ServeCLIClient/Swift/README.md

Advanced usage examples for transcription, translation, and server configuration.

```bash
# Transcribe in Spanish
swift run whisperkit-client transcribe -l es audio.wav

# Transcribe with word-level timestamps
swift run whisperkit-client transcribe --timestamp-granularities "word,segment" audio.wav

# Translate from Spanish to English
swift run whisperkit-client translate -l es audio.wav

# Use custom server
swift run whisperkit-client transcribe -s http://192.168.1.100:50060 audio.wav

# Stream transcription
swift run whisperkit-client transcribe --stream audio.wav
```

--------------------------------

### WhisperKit Client Usage Examples

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/Examples/ServeCLIClient/Python/README.md

Various command-line examples for transcribing, translating, and configuring the client with custom server settings or debug modes.

```bash
# Transcribe in Spanish
python whisperkit_client.py transcribe -l es audio.wav

# Translate to English (auto-detects source language)
python whisperkit_client.py translate audio.wav

# Use custom server and model
python whisperkit_client.py -s http://192.168.1.100:50060 -m large transcribe audio.wav

# Transcribe with word-level timestamps
python whisperkit_client.py transcribe --timestamp-granularities "word,segment" audio.wav

# Stream transcription
python whisperkit_client.py transcribe --stream audio.wav

# Debug mode to see raw JSON
python whisperkit_client.py transcribe --debug audio.wav

# Test with sample files
python whisperkit_client.py test
```

--------------------------------

### Open Xcode Project

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/BENCHMARKS.md

Open the example project in Xcode to configure environment variables for benchmarks.

```sh
xed Examples/WhisperAX
```

--------------------------------

### Install WhisperKit CLI with Homebrew

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Install the WhisperKit command-line interface application using Homebrew.

```bash
brew install whisperkit-cli
```

--------------------------------

### Start Argmax Local Server

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Starts the Argmax local server with default settings or custom host, port, model, and logging.

```bash
# Start server with default settings
BUILD_ALL=1 swift run argmax-cli serve

# Custom host and port
BUILD_ALL=1 swift run argmax-cli serve --host 0.0.0.0 --port 8080

# With specific model and verbose logging
BUILD_ALL=1 swift run argmax-cli serve --model tiny --verbose

# See all configurable parameters
BUILD_ALL=1 swift run argmax-cli serve --help
```

--------------------------------

### Install Xcode Command Line Tools

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/fastlane/README.md

Ensures the necessary Xcode command line tools are installed on the system.

```sh
xcode-select --install
```

--------------------------------

### Start WhisperKit Server

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/Examples/ServeCLIClient/Swift/README.md

Command to launch the local WhisperKit server instance.

```bash
whisperkit-cli serve
```

--------------------------------

### Python OpenAI SDK Transcription Example

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

A quick Python example demonstrating how to transcribe an audio file using the OpenAI SDK pointed at the local Argmax server.

```python
from openai import OpenAI
client = OpenAI(base_url="http://localhost:50060/v1")
result = client.audio.transcriptions.create(
    file=open("audio.wav", "rb"),
    model="tiny"  # Model parameter is required
)
print(result.text)
```

--------------------------------

### Clone Argmax OSS Swift Repository

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Clone the repository to your local machine to begin the installation process.

```bash
git clone https://github.com/argmaxinc/argmax-oss-swift.git
cd argmax-oss-swift
```

--------------------------------

### Quick Start: Transcribe Local Audio

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Initialize WhisperKit and transcribe a local audio file. Ensure the audio file path is correct.

```swift
import WhisperKit

// Initialize WhisperKit with default settings
Task {
   let pipe = try? await WhisperKit()
   let transcription = try? await pipe!.transcribe(audioPath: "path/to/your/audio.{wav,mp3,m4a,flac}")?.text
    print(transcription)
}
```

--------------------------------

### Translate Audio Files

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/Examples/ServeCLIClient/Curl/README.md

Usage examples for the translate.sh script to convert audio to English.

```bash
./translate.sh audio.wav
./translate.sh audio.wav --language es
./translate.sh audio.wav --stream true --logprobs
```

--------------------------------

### Transcribe Audio Files

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/Examples/ServeCLIClient/Curl/README.md

Usage examples for the transcribe.sh script with various optional flags.

```bash
./transcribe.sh audio.wav
./transcribe.sh audio.wav --language en --timestamp-granularities word,segment
./transcribe.sh audio.wav --stream true --logprobs
```

--------------------------------

### Initialize and Run WhisperKit Server

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/Examples/ServeCLIClient/Curl/README.md

Commands to prepare the environment and launch the local server instance.

```bash
chmod +x *.sh
```

```bash
whisperkit-cli serve --model tiny
```

--------------------------------

### Build and Test Project

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/CONTRIBUTING.md

Build the project and run tests using the Makefile. Ensure your changes are functional and adhere to project standards.

```bash
make build
make test
```

--------------------------------

### Run Swift Client

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Executes the Swift client to transcribe or translate audio files using the local server.

```bash
cd Examples/ServeCLIClient/Swift
swift run whisperkit-client transcribe audio.wav --language en
swift run whisperkit-client translate audio.wav
```

--------------------------------

### Download All Available Models

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Use this command to download all models available in the repository to your local folder.

```bash
make download-models
```

--------------------------------

### Apply Style Instructions (1.7B Model)

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

For the 1.7B model, use the `instruction` property in `GenerationOptions` to control prosody with natural language, such as speaking slowly and warmly.

```swift
var options = GenerationOptions()
options.instruction = "Speak slowly and warmly, like a storyteller."

let result = try await tts.generate(
    text: "Once upon a time...",
    speaker: .ryan,
    options: options
)
```

--------------------------------

### Build WhisperKit Client

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/Examples/ServeCLIClient/Swift/README.md

Command to compile the Swift client project.

```bash
swift build
```

--------------------------------

### Run Full Benchmarks

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/BENCHMARKS.md

Execute the full benchmark tests. This is the standard command for running all tests.

```sh
make benchmark-devices
```

--------------------------------

### Project Directory Structure

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/Examples/ServeCLIClient/Python/README.md

The file layout for the Python client implementation.

```text
Examples/ServeCLIClient/Python/
├── whisperkit_client.py    # Main CLI script with all functionality
├── test_transcribe.py      # Test script for transcription
├── test_translate.py       # Test script for translation
├── requirements.txt         # Python dependencies
├── uv.lock                 # Locked dependency versions
└── README.md               # This file
```

--------------------------------

### Download Models

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/CONTRIBUTING.md

Download required models for local testing and running the project. Specify the model size with the MODEL argument.

```bash
make download-model MODEL=tiny
```

--------------------------------

### Run WhisperKit Client Commands

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/Examples/ServeCLIClient/Swift/README.md

Basic commands for transcribing, translating, and testing audio files.

```bash
# Transcribe an audio file
swift run whisperkit-client transcribe audio.wav

# Translate an audio file to English
swift run whisperkit-client translate audio.wav

# Test with sample files
swift run whisperkit-client test
```

--------------------------------

### SpeakerKit Initialization

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Initializes SpeakerKit with default configurations, automatically downloading necessary models from HuggingFace on first run. Models are loaded lazily.

```APIDOC
## SpeakerKit()

### Description
Initializes the SpeakerKit with default PyannoteConfig. Models are downloaded from HuggingFace on first run and loaded lazily.

### Method
`SpeakerKit()`

### Parameters
None
```

--------------------------------

### List Connected Devices

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/BENCHMARKS.md

Use this command to list all connected devices and verify their connection status before running benchmarks.

```sh
make list-devices
```

--------------------------------

### Run Client Tests

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/Examples/ServeCLIClient/Python/README.md

Commands to execute the built-in test suite or individual test scripts.

```bash
# Run tests on sample files
python whisperkit_client.py test

# Or run individual test scripts
python test_transcribe.py
python test_translate.py
```

--------------------------------

### SpeakerKit Initialization with Custom Config

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Initializes SpeakerKit with a custom configuration, allowing for local model paths to be specified.

```APIDOC
## SpeakerKit(config: PyannoteConfig)

### Description
Initializes the SpeakerKit with a custom PyannoteConfig, useful for specifying local model paths and skipping the download step.

### Method
`SpeakerKit(config: PyannoteConfig)`

### Parameters
#### Path Parameters
- **config** (PyannoteConfig) - Required - Custom configuration for SpeakerKit, including model paths.
```

--------------------------------

### Run Benchmarks on Specific Devices

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/BENCHMARKS.md

Optionally specify a comma-separated list of device names to run benchmarks on. Device names can be found using 'make list-devices'.

```sh
make benchmark-devices DEVICES="iPhone 15 Pro Max,My Mac"
```

--------------------------------

### Build Argmax Local Server

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Builds the Argmax CLI with local server support using Make or manual Swift build flags.

```bash
# Build with server support
make build-local-server

# Or manually with the build flag
BUILD_ALL=1 swift build --product argmax-cli
```

--------------------------------

### Run Python OpenAI SDK Client

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Executes Python scripts to transcribe or translate audio files using the local server via the OpenAI SDK.

```bash
python whisperkit_client.py transcribe --file audio.wav --language en
python whisperkit_client.py translate --file audio.wav
```

--------------------------------

### Project Structure Overview

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/Examples/ServeCLIClient/Swift/README.md

Directory layout of the WhisperKit Swift client source code.

```text
Sources/
├── CLI.swift           # All CLI commands and client logic
└── Generated/          # Auto-generated OpenAPI client code
    ├── Client.swift
    └── Types.swift
```

--------------------------------

### Transcribe Audio File with Swift CLI

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Run the Swift CLI to transcribe an audio file. Specify the path to the downloaded model and the audio file.

```bash
swift run argmax-cli transcribe --model-path "Models/whisperkit-coreml/openai_whisper-large-v3-v20240930_626MB" --audio-path "path/to/your/audio.{wav,mp3,m4a,flac}"
```

--------------------------------

### Clone Argmax OSS Repository

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/CONTRIBUTING.md

Clone your forked repository to your local machine to begin making changes. Navigate into the cloned directory.

```bash
git clone https://github.com/[your-username]/argmax-oss-swift.git
cd argmax-oss-swift
```

--------------------------------

### Configure Playback Strategy

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Control the audio buffering behavior during real-time playback using different strategies. The `.auto` strategy dynamically adjusts buffering based on generation speed.

```swift
try await tts.play(
    text: "Long passage...",
    playbackStrategy: .auto
)
```

--------------------------------

### Clone WhisperKit Repository

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/BENCHMARKS.md

Use this command to download the WhisperKit source code for running benchmarks.

```sh
git clone git@github.com:argmaxinc/argmax-oss-swift.git
```

--------------------------------

### Diarize Audio File with SpeakerKit

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Demonstrates basic usage of SpeakerKit to diarize an audio file. It loads audio, performs diarization, and prints the detected speaker count and segments. Models are downloaded automatically on first run.

```swift
import SpeakerKit

Task {
    let speakerKit = try await SpeakerKit()

    let audioArray = try AudioProcessor.loadAudioAsFloatArray(fromPath: "audio.wav")
    let result = try await speakerKit.diarize(audioArray: audioArray)

    print("Detected \(result.speakerCount) speakers")
    for segment in result.segments {
        print(segment)
    }
}
```

--------------------------------

### Real-Time Streaming Playback

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Utilize the `play` function for real-time streaming of generated audio. This allows playback to begin before the entire audio is generated.

```swift
try await tts.play(text: "This starts playing before generation finishes.")
```

--------------------------------

### Run Curl Client Scripts

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Makes the shell scripts executable and runs them to transcribe, translate, or test audio files with the local server.

```bash
cd Examples/ServeCLIClient/Curl
chmod +x *.sh
./transcribe.sh audio.wav --language en
./translate.sh audio.wav --language es
./test.sh  # Run comprehensive test suite
```

--------------------------------

### Benchmark iOS Devices

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/fastlane/README.md

Benchmarks connected devices with configurable options.

```sh
[bundle exec] fastlane ios benchmark
```

--------------------------------

### Receive Progress Callbacks During Generation

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Implement progress callbacks to receive audio chunks as they are generated. Returning `false` from the callback will cancel the generation process early.

```swift
let result = try await tts.generate(text: "Hello!") { progress in
    print("Audio chunk: \(progress.audio.count) samples")
    if let stepTime = progress.stepTime {
        print("First step took \(stepTime)s")
    }
    return true  // return false to cancel
}
```

--------------------------------

### TTSKitExample Project Structure

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/Examples/TTS/TTSKitExample/README.md

Overview of the main files and their roles within the TTSKitExample application.

```plaintext
TTSKitExample/
├── TTSKitExampleApp.swift        entry point
├── ContentView.swift             root NavigationSplitView
├── SidebarView.swift             model management + history
├── DetailView.swift              input form, waveform, playback
├── ModelManagementView.swift     download / load / unload
├── GenerationSettingsView.swift  advanced options sheet
├── ViewModel.swift               @Observable view model
├── AudioMetadata.swift           Codable metadata embedded in .m4a files
├── WaveformView.swift            live waveform bar chart
└── ComputeUnitsView.swift        per-component compute unit picker
```

--------------------------------

### Run Debug Benchmarks

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/BENCHMARKS.md

Execute the debug benchmark tests. Set DEBUG=true to check for potential errors.

```sh
make benchmark-devices DEBUG=true
```

--------------------------------

### Upload Benchmark Results

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/fastlane/README.md

Uploads the extracted benchmark results.

```sh
[bundle exec] fastlane ios upload_results
```

--------------------------------

### Stream Audio for Transcription with Swift CLI

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Use the Swift CLI to transcribe audio streamed directly from a microphone. Ensure the model path is correctly specified.

```bash
swift run argmax-cli transcribe --model-path "Models/whisperkit-coreml/openai_whisper-large-v3-v20240930_626MB" --stream
```

--------------------------------

### Add Target Dependencies for Argmax OSS

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Specify the Argmax OSS products as dependencies for your application's target in Package.swift. You can import all kits or individual ones.

```swift
.target(
    name: "YourApp",
    dependencies: [
        // Import everything at once:
        .product(name: "ArgmaxOSS", package: "argmax-oss-swift"),

        // Or pick individual kits:
        // .product(name: "WhisperKit", package: "argmax-oss-swift"),   // speech-to-text
        // .product(name: "TTSKit", package: "argmax-oss-swift"),       // text-to-speech
        // .product(name: "SpeakerKit", package: "argmax-oss-swift"),   // speaker diarization
    ]
),
```

--------------------------------

### Control Speaker Detection with Options

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Shows how to customize speaker detection using `PyannoteDiarizationOptions`. You can specify the number of speakers, adjust the clustering threshold, and control speaker assignment exclusivity.

```swift
let audioArray = try AudioProcessor.loadAudioAsFloatArray(fromPath: "audio.wav")
let options = PyannoteDiarizationOptions(
    numberOfSpeakers: 2,               // nil = automatic detection
    clusterDistanceThreshold: 0.6,     // clustering threshold
    useExclusiveReconciliation: false   // exclusive speaker assignment per frame
)
let result = try await speakerKit.diarize(audioArray: audioArray, options: options)
```

--------------------------------

### Select Speech Decoder Mode

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Choose between `latencyOptimized` (default) for lowest time-to-first-audio or `throughputOptimized` for higher throughput at the cost of initial latency. This setting is applied when models are loaded.

```swift
// Default: latency-optimized (lowest time-to-first-audio)
let tts = try await TTSKit()
```

```swift
// Opt into throughput-optimized generation
let config = TTSKitConfig(speechDecoderMode: .throughputOptimized)
let throughputTTS = try await TTSKit(config)
```

--------------------------------

### List iOS Devices

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/fastlane/README.md

Lists all connected iOS devices.

```sh
[bundle exec] fastlane ios list_devices
```

--------------------------------

### Generate TypeScript Client from OpenAPI Spec

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Generates a TypeScript-Fetch client from the local server's OpenAPI specification using openapi-generator-cli.

```bash
# Generate TypeScript client
npx @openapitools/openapi-generator-cli generate \
  -i scripts/specs/localserver_openapi.yaml \
  -g typescript-fetch \
  -o typescript-client
```

--------------------------------

### Execute WhisperKit Scripts

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/Examples/ServeCLIClient/Curl/README.md

Basic commands for interacting with the transcription and translation scripts.

```bash
# Transcribe audio
./transcribe.sh audio.wav

# Translate audio to English
./translate.sh audio.wav --language es

# Run test suite
./test.sh
```

--------------------------------

### Generate OpenAPI Spec and Server Code

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Generates the latest OpenAPI specification and server code for the Argmax CLI using Make.

```bash
# Generate latest spec and server code
make generate-server
```

--------------------------------

### Generate Python Client from OpenAPI Spec

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Generates a Python client and types from the local server's OpenAPI specification using Swift OpenAPI Generator.

```bash
# Generate Python client
swift run swift-openapi-generator generate scripts/specs/localserver_openapi.yaml \
  --output-directory python-client \
  --mode client \
  --mode types
```

--------------------------------

### Generate Speech with TTS CLI

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Use the argmax-cli tool to generate speech from text. Options include playing audio in real-time, saving to a file, specifying language and speaker, and using text files with models and instructions.

```bash
swift run argmax-cli tts --text "Hello from the command line" --play
```

```bash
swift run argmax-cli tts --text "Save to file" --output-path output.wav
```

```bash
swift run argmax-cli tts --text "日本語テスト" --speaker ono-anna --language japanese
```

```bash
swift run argmax-cli tts --text-file article.txt --model 1.7b --instruction "Read cheerfully"
```

```bash
swift run argmax-cli tts --help
```

--------------------------------

### Generate Speech from Text

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Basic usage to generate speech from a given text. The default model is downloaded on first run and loaded lazily.

```swift
import TTSKit

Task {
    let tts = try await TTSKit()
    let result = try await tts.generate(text: "Hello from TTSKit!")
    print("Generated \(result.audioDuration)s of audio at \(result.sampleRate)Hz")
}
```

--------------------------------

### Select TTSKit Model Size

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Choose between different model sizes for TTSKit. The smaller model is faster and compatible with all platforms, while the larger model offers higher quality on macOS and supports style instructions.

```swift
// Fast, runs on all platforms (~1 GB download)
let tts = try await TTSKit(TTSKitConfig(model: .qwen3TTS_0_6b))
```

```swift
// Higher quality, macOS only (~2.2 GB download, supports style instructions)
let tts = try await TTSKit(TTSKitConfig(model: .qwen3TTS_1_7b))
```

--------------------------------

### iOS Fastlane Actions

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/fastlane/README.md

A collection of Fastlane actions for managing iOS devices and benchmark results.

```APIDOC
## ios list_devices

### Description
List all connected iOS devices.

### Command
`fastlane ios list_devices`

## ios benchmark

### Description
Benchmark connected iOS devices with configurable options.

### Command
`fastlane ios benchmark`

## ios extract_results

### Description
Extract benchmark results from the iOS environment.

### Command
`fastlane ios extract_results`

## ios upload_results

### Description
Upload the extracted benchmark results.

### Command
`fastlane ios upload_results`
```

--------------------------------

### BibTeX Citation for Argmax OSS

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Use this BibTeX entry when citing the Argmax OSS SDK, including WhisperKit, SpeakerKit, and TTSKit, in academic work.

```bibtex
@misc{whisperkit-argmax,
   title = {Argmax OSS: On-device Speech AI with WhisperKit, SpeakerKit and TTSKit},
   author = {Argmax, Inc.},
   year = {2024},
   URL = {https://github.com/argmaxinc/argmax-oss-swift}
}
```

--------------------------------

### Use Local SpeakerKit Models

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Configure SpeakerKit to use local CoreML models by specifying the model folder path. This bypasses the automatic download from HuggingFace.

```swift
let config = PyannoteConfig(modelFolder: "/path/to/models")
let speakerKit = try await SpeakerKit(config)
```

--------------------------------

### Transcribe Audio with Diarization using argmax-cli

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Combines transcription and diarization using the argmax-cli tool. This command processes an audio file and outputs speaker-attributed transcriptions.

```bash
swift run argmax-cli transcribe --audio-path audio.wav --diarization
```

--------------------------------

### POST /v1/audio/transcriptions

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Transcribes audio to text using the local server. Supports various audio formats and includes options for model selection, language detection, and streaming.

```APIDOC
## POST /v1/audio/transcriptions

### Description
Transcribes audio to text. This endpoint allows users to send audio files and receive text transcriptions.

### Method
POST

### Endpoint
/v1/audio/transcriptions

### Parameters
#### Path Parameters
None

#### Query Parameters
None

#### Request Body
- **file** (Audio file) - Required - Audio file (wav, mp3, m4a, flac)
- **model** (string) - Server default - Model identifier
- **language** (string) - Auto-detect - Source language code
- **prompt** (string) - None - Text to guide transcription
- **response_format** (string) - verbose_json - Output format (json, verbose_json)
- **temperature** (number) - 0.0 - Sampling temperature (0.0-1.0)
- **timestamp_granularities[]** (string) - segment - Timing detail (word, segment)
- **stream** (boolean) - false - Enable streaming
```

--------------------------------

### Extract Benchmark Results

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/fastlane/README.md

Extracts the results from the benchmark process.

```sh
[bundle exec] fastlane ios extract_results
```

--------------------------------

### Generate Custom CoreML Models

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Initialize WhisperKit with a custom model and model repository. This is used for deploying fine-tuned models.

```swift
let config = WhisperKitConfig(model: "large-v3-v20240930_626MB", modelRepo: "username/your-model-repo")
let pipe = try? await WhisperKit(config)
```

--------------------------------

### Configure TTSKit with a Private Model Token

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/Examples/TTS/TTSKitExample/README.md

Set a Hugging Face model token for private model forks. This configuration is used when initializing TTSKit.

```swift
let config = TTSKitConfig(model: selectedPreset, modelToken: "hf_...")
```

--------------------------------

### Perform Diarization with argmax-cli

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Use the argmax-cli tool for standalone speaker diarization. Options include verbose output, saving RTTM results, and specifying the number of speakers.

```bash
swift run argmax-cli diarize --audio-path audio.wav --verbose
```

```bash
swift run argmax-cli diarize --audio-path audio.wav --rttm-path output.rttm
```

```bash
swift run argmax-cli diarize --audio-path audio.wav --num-speakers 3
```

```bash
swift run argmax-cli diarize --help
```

--------------------------------

### POST /v1/audio/translations

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Translates audio to English using the local server. Supports various audio formats and includes options for model selection, language detection, and streaming.

```APIDOC
## POST /v1/audio/translations

### Description
Translates audio to English. This endpoint allows users to send audio files and receive English translations.

### Method
POST

### Endpoint
/v1/audio/translations

### Parameters
#### Path Parameters
None

#### Query Parameters
None

#### Request Body
- **file** (Audio file) - Required - Audio file (wav, mp3, m4a, flac)
- **model** (string) - Server default - Model identifier
- **language** (string) - Auto-detect - Source language code
- **prompt** (string) - None - Text to guide transcription
- **response_format** (string) - verbose_json - Output format (json, verbose_json)
- **temperature** (number) - 0.0 - Sampling temperature (0.0-1.0)
- **timestamp_granularities[]** (string) - segment - Timing detail (word, segment)
- **stream** (boolean) - false - Enable streaming
```

--------------------------------

### Regenerate Client Code

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/Examples/ServeCLIClient/Swift/README.md

Script to update generated client code based on the OpenAPI specification.

```bash
./updateClient.sh
```

--------------------------------

### Add Swift Package Dependency

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Add the Argmax OSS Swift package to your project's dependencies in Package.swift.

```swift
dependencies: [
    .package(url: "https://github.com/argmaxinc/argmax-oss-swift.git", from: "0.9.0"),
],
```

--------------------------------

### Customize Generation Options

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Adjust generation parameters such as temperature, topK, repetition penalty, and chunking strategy. The `concurrentWorkerCount` can be set to `nil` to use a default suitable for the device.

```swift
// Defaults recommended by Qwen
var options = GenerationOptions()
options.temperature = 0.9
options.topK = 50
options.repetitionPenalty = 1.05
options.maxNewTokens = 245

// Long text is automatically split at sentence boundaries
options.chunkingStrategy = .sentence
options.concurrentWorkerCount = nil  // nil = all chunks run concurrently with a good default for the device

let result = try await tts.generate(text: longArticle, options: options)
```

--------------------------------

### Commit Changes

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/CONTRIBUTING.md

Commit your changes with a clear and descriptive message. This helps in tracking modifications and understanding the purpose of each commit.

```bash
git commit -am "Add a new feature"
```

--------------------------------

### Push Changes to Fork

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/CONTRIBUTING.md

Push your committed changes to your personal fork on GitHub. This makes your changes available for creating a pull request.

```bash
git push origin my-branch
```

--------------------------------

### Generate Speech with Custom Voice and Language

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Customize the speech output by selecting a specific voice and language. A wide range of built-in voices and languages are supported.

```swift
let result = try await tts.generate(
    text: "こんにちは世界",
    speaker: .onoAnna,
    language: .japanese
)
```

--------------------------------

### Specify WhisperKit Model

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Initialize WhisperKit with a specific model. The model name must exist in the HuggingFace repo.

```swift
let pipe = try? await WhisperKit(WhisperKitConfig(model: "large-v3-v20240930_626MB"))
```

--------------------------------

### Diarize Audio Array with Options

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Performs speaker diarization with custom options to control parameters like the number of speakers and clustering thresholds.

```APIDOC
## diarize(audioArray: [Float], options: PyannoteDiarizationOptions) async throws -> SpeakerDiarizationResult

### Description
Performs speaker diarization with specified options to fine-tune the detection process, such as setting the number of speakers or adjusting clustering thresholds.

### Method
`diarize(audioArray: [Float], options: PyannoteDiarizationOptions) async throws`

### Parameters
#### Path Parameters
- **audioArray** ([Float]) - Required - An array of floating-point numbers representing the audio data.
- **options** (PyannoteDiarizationOptions) - Required - Options to customize diarization, including `numberOfSpeakers`, `clusterDistanceThreshold`, and `useExclusiveReconciliation`.

### Response
#### Success Response
- **speakerCount** (Int) - The total number of unique speakers detected.
- **segments** ([SpeakerSegment]) - An array of SpeakerSegment objects, each containing start time, end time, and speaker label.
```

--------------------------------

### Save Generated Audio

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Save the generated speech audio to a file in WAV or M4A format. The output directory is determined by `FileManager.default`.

```swift
let result = try await tts.generate(text: "Save me!")
let outputDir = FileManager.default.urls(for: .documentDirectory, in: .userDomainMask)[0]

// Save as .wav or .m4a (AAC)
try await AudioOutput.saveAudio(result.audio, toFolder: outputDir, filename: "output", format: .m4a)
```

--------------------------------

### Combine Diarization with Transcription

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Integrates SpeakerKit with WhisperKit to attribute transcribed text to specific speakers. It processes audio, performs transcription and diarization, and then merges the results.

```swift
import WhisperKit
import SpeakerKit

let whisperKit = try await WhisperKit()
let speakerKit = try await SpeakerKit()

let audioArray = try AudioProcessor.loadAudioAsFloatArray(fromPath: "audio.wav")
let transcription = try await whisperKit.transcribe(audioArray: audioArray)
let diarization = try await speakerKit.diarize(audioArray: audioArray)

let speakerSegments = diarization.addSpeakerInfo(to: transcription)

for group in speakerSegments {
    for segment in group {
        print("\(segment.speaker): \(segment.text)")
    }
}
```

--------------------------------

### Diarize Audio Array

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Performs speaker diarization on a given audio array using the default options.

```APIDOC
## diarize(audioArray: [Float]) async throws -> SpeakerDiarizationResult

### Description
Performs speaker diarization on the provided audio data. The result includes speaker count and time-stamped segments.

### Method
`diarize(audioArray: [Float]) async throws`

### Parameters
#### Path Parameters
- **audioArray** ([Float]) - Required - An array of floating-point numbers representing the audio data.

### Response
#### Success Response
- **speakerCount** (Int) - The total number of unique speakers detected.
- **segments** ([SpeakerSegment]) - An array of SpeakerSegment objects, each containing start time, end time, and speaker label.
```

--------------------------------

### Generate RTTM Output with SpeakerKit

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Generates output in the RTTM (Rich Transcription Time Model) format from SpeakerKit diarization results. This format is commonly used for speech processing evaluation.

```swift
let speakerKit = try await SpeakerKit()

let audioArray = try AudioProcessor.loadAudioAsFloatArray(fromPath: "meeting.wav")
let diarization = try await speakerKit.diarize(audioArray: audioArray)

let rttmLines = SpeakerKit.generateRTTM(from: diarization, fileName: "meeting")
for line in rttmLines {
    print(line)
}
```

--------------------------------

### Generate RTTM Output

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Generates output in the RTTM (Rich Transcription Time Marked) format from diarization results.

```APIDOC
## generateRTTM(from: SpeakerDiarizationResult, fileName: String) -> [String]

### Description
Generates a list of strings formatted as RTTM lines, representing the diarization results for a given audio file.

### Method
`generateRTTM(from: SpeakerDiarizationResult, fileName: String)`

### Parameters
#### Path Parameters
- **diarization** (SpeakerDiarizationResult) - Required - The result object from the `diarize` method.
- **fileName** (String) - Required - The base name for the RTTM output.

### Response
#### Success Response
- **rttmLines** ([String]) - An array of strings, each representing a line in the RTTM format.
```

--------------------------------

### Add Speaker Info to Transcription

Source: https://github.com/argmaxinc/argmax-oss-swift/blob/main/README.md

Merges diarization results with transcription data to attribute speech segments to specific speakers.

```APIDOC
## addSpeakerInfo(to: Transcription) -> [[SpeakerSegment]]

### Description
Combines speaker diarization results with WhisperKit transcriptions to create speaker-attributed text segments. Supports `.subsegment` (default) and `.segment` matching strategies.

### Method
`addSpeakerInfo(to: Transcription)`

### Parameters
#### Path Parameters
- **transcription** (Transcription) - Required - The transcription result from WhisperKit.

### Response
#### Success Response
- **speakerSegments** ([[SpeakerSegment]]) - A nested array where each inner array represents segments attributed to a single speaker.
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.