lattifai 0.1.4__py3-none-any.whl → 0.2.0__py3-none-any.whl

lattifai/__init__.py

@@ -1,3 +1,7 @@
+import os
+import sys
+import warnings
+
 from .base_client import LattifAIError
 from .io import SubtitleIO
 
@@ -13,6 +17,30 @@ except Exception:
     __version__ = '0.1.0'  # fallback version
 
 
+# Check and auto-install k2 if not present
+def _check_and_install_k2():
+    """Check if k2 is installed and attempt to install it if not."""
+    try:
+        import k2
+    except ImportError:
+        import subprocess
+
+        print('k2 is not installed. Attempting to install k2...')
+        try:
+            subprocess.check_call([sys.executable, '-m', 'pip', 'install', 'install-k2'])
+            subprocess.check_call([sys.executable, '-m', 'install_k2'])
+            import k2  # Try importing again after installation
+
+            print('k2 installed successfully.')
+        except Exception as e:
+            warnings.warn(f'Failed to install k2 automatically. Please install it manually. Error: {e}')
+    return True
+
+
+# Auto-install k2 on first import
+_check_and_install_k2()
+
+
 # Lazy import for LattifAI to avoid dependency issues during basic import
 def __getattr__(name):
     if name == 'LattifAI':

lattifai/bin/align.py

@@ -13,6 +13,16 @@ from lattifai.bin.cli_base import cli
     default='auto',
     help='Input Subtitle format.',
 )
+@click.option(
+    '-D',
+    '--device',
+    type=click.Choice(['cpu', 'cuda', 'mps'], case_sensitive=False),
+    default='cpu',
+    help='Device to use for inference.',
+)
+@click.option(
+    '-M', '--model_name_or_path', type=str, default='Lattifai/Lattice-1-Alpha', help='Lattifai model name or path'
+)
 @click.argument(
     'input_audio_path',
     type=click.Path(exists=True, dir_okay=False),
@@ -30,13 +40,15 @@ def align(
     input_subtitle_path: Pathlike,
     output_subtitle_path: Pathlike,
     input_format: str = 'auto',
+    device: str = 'cpu',
+    model_name_or_path: str = 'Lattifai/Lattice-1-Alpha',
 ):
     """
     Command used to align audio with subtitles
     """
     from lattifai import LattifAI
 
-    client = LattifAI()
+    client = LattifAI(model_name_or_path=model_name_or_path, device=device)
     client.alignment(
-        input_audio_path, input_subtitle_path, format=input_format, output_subtitle_path=output_subtitle_path
+        input_audio_path, input_subtitle_path, format=input_format.lower(), output_subtitle_path=output_subtitle_path
     )

lattifai/bin/cli_base.py

@@ -8,6 +8,11 @@ def cli():
     """
     The shell entry point to Lattifai, a tool for audio data manipulation.
     """
+    # Load environment variables from .env file
+    from dotenv import load_dotenv
+
+    load_dotenv()
+
     logging.basicConfig(
         format='%(asctime)s %(levelname)s [%(filename)s:%(lineno)d] %(message)s',
         level=logging.INFO,

lattifai/client.py

@@ -24,9 +24,10 @@ class LattifAI(SyncAPIClient):
         self,
         *,
         api_key: Optional[str] = None,
+        model_name_or_path: str = 'Lattifai/Lattice-1-Alpha',
+        device: Optional[str] = None,
         base_url: Optional[str] = None,
-        device: str = 'cpu',
-        timeout: Union[float, int] = 60.0,
+        timeout: Union[float, int] = 120.0,
         max_retries: int = 2,
         default_headers: Optional[Dict[str, str]] = None,
     ) -> None:
@@ -52,19 +53,31 @@ class LattifAI(SyncAPIClient):
         )
 
         # Initialize components
-        model_name_or_path = '/Users/feiteng/GEEK/OmniCaptions/HF_models/Lattice-1-Alpha'
-
         if not Path(model_name_or_path).exists():
-            from huggingface_hub import hf_hub_download
-
-            model_path = hf_hub_download(repo_id=model_name_or_path, repo_type='model')
+            from huggingface_hub import snapshot_download
+            from huggingface_hub.errors import LocalEntryNotFoundError
+
+            try:
+                model_path = snapshot_download(repo_id=model_name_or_path, repo_type='model')
+            except LocalEntryNotFoundError:
+                os.environ['HF_ENDPOINT'] = 'https://hf-mirror.com'
+                model_path = snapshot_download(repo_id=model_name_or_path, repo_type='model')
         else:
             model_path = model_name_or_path
 
+        # device setup
+        if device is None:
+            import torch
+
+            device = 'cpu'
+            if torch.backends.mps.is_available():
+                device = 'mps'
+            elif torch.cuda.is_available():
+                device = 'cuda'
+
         self.tokenizer = LatticeTokenizer.from_pretrained(
             client_wrapper=self,
-            model_path=f'{model_path}/words.bin',
-            g2p_model_path=f'{model_path}/g2p.bin' if Path(f'{model_path}/g2p.bin').exists() else None,
+            model_path=model_path,
             device=device,
         )
         self.worker = Lattice1AlphaWorker(model_path, device=device, num_threads=8)
@@ -119,13 +132,10 @@ if __name__ == '__main__':
     import sys
 
     if len(sys.argv) == 4:
-        pass
+        audio, subtitle, output = sys.argv[1:]
     else:
         audio = 'tests/data/SA1.wav'
-        text = 'tests/data/SA1.TXT'
-
-    alignments = client.alignment(audio, text)
-    print(alignments)
+        subtitle = 'tests/data/SA1.TXT'
+        output = None
 
-    alignments = client.alignment(audio, 'not paired texttttt', format='txt')
-    print(alignments)
+    alignments = client.alignment(audio, subtitle, output_subtitle_path=output)

lattifai/tokenizers/phonemizer.py

@@ -1,7 +1,7 @@
 import re
 from typing import List, Optional, Union
 
-from dp.phonemizer import Phonemizer
+from dp.phonemizer import Phonemizer  # g2p-phonemizer
 from num2words import num2words
 
 LANGUAGE = 'omni'

lattifai/tokenizers/tokenizer.py

@@ -33,22 +33,26 @@ class LatticeTokenizer:
     def from_pretrained(
         client_wrapper: SyncAPIClient,
         model_path: str,
-        g2p_model_path: Optional[str] = None,
         device: str = 'cpu',
         compressed: bool = True,
     ):
         """Load tokenizer from exported binary file"""
+        from pathlib import Path
+
+        words_model_path = f'{model_path}/words.bin'
         if compressed:
-            with gzip.open(model_path, 'rb') as f:
+            with gzip.open(words_model_path, 'rb') as f:
                 data = pickle.load(f)
         else:
-            with open(model_path, 'rb') as f:
+            with open(words_model_path, 'rb') as f:
                 data = pickle.load(f)
 
         tokenizer = LatticeTokenizer(client_wrapper=client_wrapper)
         tokenizer.words = data['words']
         tokenizer.dictionaries = defaultdict(list, data['dictionaries'])
         tokenizer.oov_word = data['oov_word']
+
+        g2p_model_path = f'{model_path}/g2p.bin' if Path(f'{model_path}/g2p.bin').exists() else None
         if g2p_model_path:
             tokenizer.g2p_model = G2Phonemizer(g2p_model_path, device=device)
         return tokenizer

lattifai/workers/lattice1_alpha.py

@@ -5,8 +5,9 @@ from typing import Any, BinaryIO, Dict, Tuple, Union
 
 import numpy as np
 import onnxruntime as ort
+import resampy
+import soundfile as sf
 import torch
-import torchaudio
 from lhotse import FbankConfig
 from lhotse.features.kaldi.layers import Wav2LogFilterBank
 from lhotse.utils import Pathlike
@@ -16,8 +17,6 @@ class Lattice1AlphaWorker:
     """Worker for processing audio with LatticeGraph."""
 
     def __init__(self, model_path: Pathlike, device: str = 'cpu', num_threads: int = 8) -> None:
-        if device != 'cpu':
-            raise NotImplementedError(f'Only cpu is supported for now, got device={device}.')
         self.config = json.load(open(f'{model_path}/config.json'))
 
         # SessionOptions
@@ -28,8 +27,11 @@ class Lattice1AlphaWorker:
         sess_options.add_session_config_entry('session.intra_op.allow_spinning', '0')
 
         providers = []
-        if device.startswith('cuda') or ort.get_all_providers().count('CUDAExecutionProvider') > 0:
+        if device.startswith('cuda') and ort.get_all_providers().count('CUDAExecutionProvider') > 0:
             providers.append('CUDAExecutionProvider')
+        elif device.startswith('mps') and ort.get_all_providers().count('MPSExecutionProvider') > 0:
+            providers.append('MPSExecutionProvider')
+
         self.acoustic_ort = ort.InferenceSession(
             f'{model_path}/acoustic_opt.onnx',
             sess_options,
@@ -48,21 +50,38 @@ class Lattice1AlphaWorker:
         _start = time.time()
         # audio -> features -> emission
         features = self.extractor(audio)  # (1, T, D)
-        ort_inputs = {
-            'features': features.cpu().numpy(),
-            'feature_lengths': np.array([features.size(1)], dtype=np.int64),
-        }
-        emission = self.acoustic_ort.run(None, ort_inputs)[0]  # (1, T, vocab_size) numpy
+        if features.shape[1] > 6000:
+            features_list = torch.split(features, 6000, dim=1)
+            emissions = []
+            for features in features_list:
+                ort_inputs = {
+                    'features': features.cpu().numpy(),
+                    'feature_lengths': np.array([features.size(1)], dtype=np.int64),
+                }
+                emission = self.acoustic_ort.run(None, ort_inputs)[0]  # (1, T, vocab_size) numpy
+                emissions.append(emission)
+            emission = torch.cat(
+                [torch.from_numpy(emission).to(self.device) for emission in emissions], dim=1
+            )  # (1, T, vocab_size)
+        else:
+            ort_inputs = {
+                'features': features.cpu().numpy(),
+                'feature_lengths': np.array([features.size(1)], dtype=np.int64),
+            }
+            emission = self.acoustic_ort.run(None, ort_inputs)[0]  # (1, T, vocab_size) numpy
+            emission = torch.from_numpy(emission).to(self.device)
+
         self.timings['emission'] += time.time() - _start
-        return torch.from_numpy(emission).to(self.device)  # (1, T, vocab_size) torch
+        return emission  # (1, T, vocab_size) torch
 
     def load_audio(self, audio: Union[Pathlike, BinaryIO]) -> Tuple[torch.Tensor, int]:
         # load audio
-        waveform, sample_rate = torchaudio.load(audio, channels_first=True)
-        if waveform.size(0) > 1:  # TODO: support choose channel
-            waveform = torch.mean(waveform, dim=0, keepdim=True)
+        waveform, sample_rate = sf.read(audio, always_2d=True, dtype='float32')
+        if waveform.shape[1] > 1:  # TODO: support choose channel
+            waveform = np.mean(waveform, axis=1, keepdims=True)
         if sample_rate != self.config['sample_rate']:
-            waveform = torchaudio.functional.resample(waveform, sample_rate, self.config['sample_rate'])
+            waveform = resampy.resample(waveform, sample_rate, self.config['sample_rate'], axis=0)
+        waveform = torch.from_numpy(waveform.T).to(self.device)  # (1, L)
         return waveform
 
     def alignment(
@@ -102,9 +121,14 @@ class Lattice1AlphaWorker:
         self.timings['decoding_graph'] += time.time() - _start
 
         _start = time.time()
+        if self.device.type == 'mps':
+            device = 'cpu'  # k2 does not support mps yet
+        else:
+            device = self.device
+
         results, labels = align_segments(
-            emission.to(self.device) * acoustic_scale,
-            decoding_graph.to(self.device),
+            emission.to(device) * acoustic_scale,
+            decoding_graph.to(device),
             torch.tensor([emission.shape[1]], dtype=torch.int32),
             search_beam=100,
             output_beam=40,

lattifai-0.2.0.dist-info/METADATA

@@ -0,0 +1,294 @@
+Metadata-Version: 2.4
+Name: lattifai
+Version: 0.2.0
+Summary: Lattifai Python SDK: Seamless Integration with Lattifai's Speech and Video AI Services
+Author-email: Lattifai Technologies <tech@lattifai.com>
+Maintainer-email: Lattice <tech@lattifai.com>
+License: MIT License
+        
+        Copyright (c) 2025 Lattifai.
+        
+        Permission is hereby granted, free of charge, to any person obtaining a copy
+        of this software and associated documentation files (the "Software"), to deal
+        in the Software without restriction, including without limitation the rights
+        to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+        copies of the Software, and to permit persons to whom the Software is
+        furnished to do so, subject to the following conditions:
+        
+        The above copyright notice and this permission notice shall be included in all
+        copies or substantial portions of the Software.
+        
+        THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+        IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+        FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+        AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+        LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+        OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+        SOFTWARE.
+Project-URL: Homepage, https://github.com/lattifai/lattifai-python
+Project-URL: Documentation, https://github.com/lattifai/lattifai-python/README.md
+Project-URL: Bug Tracker, https://github.com/lattifai/lattifai-python/issues
+Project-URL: Discussions, https://github.com/lattifai/lattifai-python/discussions
+Project-URL: Changelog, https://github.com/lattifai/lattifai-python/CHANGELOG.md
+Keywords: lattifai,speech recognition,video analysis,ai,sdk,api client
+Classifier: Development Status :: 5 - Production/Stable
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: Science/Research
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Operating System :: MacOS :: MacOS X
+Classifier: Operating System :: POSIX :: Linux
+Classifier: Operating System :: Microsoft :: Windows
+Classifier: Topic :: Multimedia :: Sound/Audio
+Classifier: Topic :: Multimedia :: Video
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: lattifai-core>=0.2.0
+Requires-Dist: httpx
+Requires-Dist: python-dotenv
+Requires-Dist: lhotse>=1.26.0
+Requires-Dist: colorful>=0.5.6
+Requires-Dist: pysubs2
+Requires-Dist: praatio
+Requires-Dist: tgt
+Requires-Dist: onnxruntime
+Requires-Dist: resampy
+Requires-Dist: g2p-phonemizer==0.1.1
+Provides-Extra: numpy
+Requires-Dist: numpy; extra == "numpy"
+Provides-Extra: test
+Requires-Dist: pytest; extra == "test"
+Requires-Dist: pytest-cov; extra == "test"
+Requires-Dist: ruff; extra == "test"
+Requires-Dist: numpy; extra == "test"
+Provides-Extra: all
+Requires-Dist: numpy; extra == "all"
+Requires-Dist: pytest; extra == "all"
+Requires-Dist: pytest-cov; extra == "all"
+Requires-Dist: ruff; extra == "all"
+Dynamic: license-file
+
+# LattifAI Python
+
+[![PyPI version](https://badge.fury.io/py/lattifai.svg)](https://badge.fury.io/py/lattifai)
+
+<p align="center">
+   🌐 <a href="https://lattifai.com"><b>Official Website</b></a> &nbsp&nbsp | &nbsp&nbsp 🖥️ <a href="https://github.com/lattifai/lattifai-python">GitHub</a> &nbsp&nbsp | &nbsp&nbsp 🤗 <a href="https://huggingface.co/Lattifai/Lattice-1-Alpha">Model</a> &nbsp&nbsp | &nbsp&nbsp 📑 <a href="https://lattifai.com/blogs">Blog</a> &nbsp&nbsp | &nbsp&nbsp <a href="https://discord.gg/gTZqdaBJ"><img src="https://img.shields.io/badge/Discord-Join-5865F2?logo=discord&logoColor=white" alt="Discord" style="vertical-align: middle;"></a>
+</p>
+
+Advanced forced alignment and subtitle generation powered by [Lattice-1-Alpha](https://huggingface.co/Lattifai/Lattice-1-Alpha) model.
+
+## Installation
+
+```bash
+pip install install-k2
+# The installation will automatically detect and use your already installed PyTorch version.
+install-k2  # Install k2
+
+pip install lattifai
+```
+
+> **⚠️ Important**: You must run `install-k2` before using the lattifai library.
+
+## Quick Start
+
+### Command Line
+
+```bash
+# Align audio with subtitle
+lattifai align audio.wav subtitle.srt output.srt
+
+# Convert subtitle format
+lattifai subtitle convert input.srt output.vtt
+```
+#### lattifai align options
+```
+> lattifai align --help
+Usage: lattifai align [OPTIONS] INPUT_AUDIO_PATH INPUT_SUBTITLE_PATH OUTPUT_SUBTITLE_PATH
+
+  Command used to align audio with subtitles
+
+Options:
+  -F, --input_format [srt|vtt|ass|txt|auto]  Input Subtitle format.
+  -D, --device [cpu|cuda|mps]                Device to use for inference.
+  --help                                     Show this message and exit.
+```
+
+### Python API
+
+```python
+from lattifai import LattifAI
+
+# Initialize client
+client = LattifAI(
+    api_key: Optional[str] = None,
+    model_name_or_path='Lattifai/Lattice-1-Alpha',
+    device='cpu',  # 'cpu', 'cuda', or 'mps'
+)
+
+# Perform alignment
+result = client.alignment(
+    audio="audio.wav",
+    subtitle="subtitle.srt",
+    output_subtitle_path="output.srt"
+)
+```
+
+## Supported Formats
+
+**Audio**: WAV, MP3, FLAC, M4A, OGG
+**Subtitle**: SRT, VTT, ASS, TXT (plain text)
+
+## API Reference
+
+### LattifAI
+
+```python
+LattifAI(
+    api_key: Optional[str] = None,
+    model_name_or_path: str = 'Lattifai/Lattice-1-Alpha',
+    device: str = 'cpu'  # 'cpu', 'cuda', or 'mps'
+)
+```
+
+### alignment()
+
+```python
+client.alignment(
+    audio: str,                    # Path to audio file
+    subtitle: str,                 # Path to subtitle/text file
+    format: Optional[str] = None,  # 'srt', 'vtt', 'ass', 'txt' (auto-detect if None)
+    output_subtitle_path: Optional[str] = None
+) -> str
+```
+
+## Examples
+
+### Basic Text Alignment
+
+```python
+client = LattifAI()
+client.alignment(
+    audio="speech.wav",
+    subtitle="transcript.txt",
+    format="txt",
+    output_subtitle_path="output.srt"
+)
+```
+
+### Batch Processing
+
+```python
+from pathlib import Path
+
+client = LattifAI()
+audio_dir = Path("audio_files")
+subtitle_dir = Path("subtitles")
+output_dir = Path("aligned")
+
+for audio in audio_dir.glob("*.wav"):
+    subtitle = subtitle_dir / f"{audio.stem}.srt"
+    if subtitle.exists():
+        client.alignment(
+            audio=audio,
+            subtitle=subtitle,
+            output_subtitle_path=output_dir / f"{audio.stem}_aligned.srt"
+        )
+```
+
+### GPU Acceleration
+
+```python
+# NVIDIA GPU
+client = LattifAI(device='cuda')
+
+# Apple Silicon
+client = LattifAI(device='mps')
+
+# CLI
+lattifai align --device mps audio.wav subtitle.srt output.srt
+```
+
+## Configuration
+
+### API Key Setup
+
+First, create your API key at [https://lattifai.com/dashboard/api-keys](https://lattifai.com/dashboard/api-keys)
+
+**Recommended: Using .env file**
+
+Create a `.env` file in your project root:
+```bash
+LATTIFAI_API_KEY=your-api-key
+```
+
+The library automatically loads the `.env` file (python-dotenv is included as a dependency).
+
+**Alternative: Environment variable**
+```bash
+export LATTIFAI_API_KEY="your-api-key"
+```
+
+## Model Information
+
+**[Lattice-1-Alpha](https://huggingface.co/Lattifai/Lattice-1-Alpha)** features:
+- State-of-the-art alignment precision
+- **Language Support**: Currently supports English only. The upcoming **Lattice-1** release will support English, Chinese, and mixed English-Chinese content.
+- Handles noisy audio and imperfect transcripts
+- Optimized for CPU and GPU (CUDA/MPS)
+
+**Requirements**:
+- Python 3.9+
+- 4GB RAM recommended
+- ~2GB storage for model files
+
+## Development
+
+### Setup
+
+```bash
+git clone https://github.com/lattifai/lattifai-python.git
+cd lattifai-python
+pip install -e ".[test]"
+./scripts/install-hooks.sh  # Optional: install pre-commit hooks
+```
+
+### Testing
+
+```bash
+pytest                        # Run all tests
+pytest --cov=src             # With coverage
+pytest tests/test_basic.py   # Specific test
+```
+
+### Code Quality
+
+```bash
+ruff check src/ tests/       # Lint
+ruff format src/ tests/      # Format
+isort src/ tests/            # Sort imports
+```
+
+## Contributing
+
+1. Fork the repository
+2. Create a feature branch
+3. Make changes and add tests
+4. Run `pytest` and `ruff check`
+5. Submit a pull request
+
+## License
+
+Apache License 2.0
+
+## Support
+
+- **Issues**: [GitHub Issues](https://github.com/lattifai/lattifai-python/issues)
+- **Discussions**: [GitHub Discussions](https://github.com/lattifai/lattifai-python/discussions)
+- **Discord**: [Join our community](https://discord.gg/gTZqdaBJ)

lattifai-0.2.0.dist-info/RECORD

@@ -0,0 +1,22 @@
+lattifai/__init__.py,sha256=JXUg0dT74UyAtKOjewRs9ijr5sl9SYsc6oU_WItY314,1497
+lattifai/base_client.py,sha256=ktFtATjL9pLSJUD-VqeJKA1FHkrsGHX7Uq_x00H7gO8,3322
+lattifai/client.py,sha256=NjHUqMZFoRxuyxdzBNEcn5kU57gJl31FSb6i9DDcoCw,5059
+lattifai/bin/__init__.py,sha256=7YhmtEM8kbxJtz2-KIskvpLKBZAvkMSceVx8z4fkgQ4,61
+lattifai/bin/align.py,sha256=nw-wABsNyxC8zN3siiqgNi1Foou3cYhVzIAomuVrFOY,1436
+lattifai/bin/cli_base.py,sha256=y535WXDRX8StloFn9icpfw7nQt0JxuWBIuPMnRxAYy8,392
+lattifai/bin/subtitle.py,sha256=bUWImAHpvyY59Vskqb5loQiD5ytQOxR8lTQRiQ4LyNA,647
+lattifai/io/__init__.py,sha256=vHWRN7MvAch-GUeFqqO-gM57SM-4YOpGUjIxFJdjfPA,671
+lattifai/io/reader.py,sha256=ErPnPMUvYQpjZ7Vd86EsHUkOcEfKdoI8iM3yKHRzSOQ,2576
+lattifai/io/supervision.py,sha256=5UfSsgBhXoDU3-6drDtoD7y8HIiA4xRKZnbOKgeejwM,354
+lattifai/io/writer.py,sha256=1eAEFLlL8kricxRDPFBtVmeC4IiFyFnjbWXvw0VU-q4,2036
+lattifai/tokenizers/__init__.py,sha256=aqv44PDtq6g3oFFKW_l4HSR5ywT5W8eP1dHHywIvBfs,72
+lattifai/tokenizers/phonemizer.py,sha256=SfRi1KIMpmaao6OVmR1h_I_3QU-vrE6D5bh72Afg5XM,1759
+lattifai/tokenizers/tokenizer.py,sha256=u4lgS6-jN9cLuMNIojA4Swfsqb1EcyI7Bh_iw7tuL-s,5818
+lattifai/workers/__init__.py,sha256=s6YfkIq4FDIAzY9sPjRpXnJfszj2repqnMTqydRM5Zw,83
+lattifai/workers/lattice1_alpha.py,sha256=1VFo59EcygEctTHOhkcII8v3_mrj8JEJ8Fcaqk_7LVo,5762
+lattifai-0.2.0.dist-info/licenses/LICENSE,sha256=LNuoH5jpXXNKgjQ3XLwztFq8D3O7kZI-LSg81o4ym2M,1065
+lattifai-0.2.0.dist-info/METADATA,sha256=bXb6z5D1C-9YwHeycSFs8SAhUp8VNJbE9u-J9lvYMZ8,8997
+lattifai-0.2.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+lattifai-0.2.0.dist-info/entry_points.txt,sha256=CwTI2NbJvF9msIHboAfTA99cmDr_HOWoODjS8R64JOw,131
+lattifai-0.2.0.dist-info/top_level.txt,sha256=-OVWZ68YYFcTN13ARkLasp2OUappe9wEVq-CKes7jM4,17
+lattifai-0.2.0.dist-info/RECORD,,

{lattifai-0.1.4.dist-info → lattifai-0.2.0.dist-info}/entry_points.txt

@@ -1,3 +1,4 @@
 [console_scripts]
+install-k2 = scripts.install_k2:install_k2
 lattifai = lattifai.bin:cli
 lattifai-align = lattifai.bin.align:align

{lattifai-0.1.4.dist-info → lattifai-0.2.0.dist-info}/top_level.txt

@@ -1 +1,2 @@
 lattifai
+scripts

lattifai-0.1.4.dist-info/METADATA

@@ -1,467 +0,0 @@
-Metadata-Version: 2.4
-Name: lattifai
-Version: 0.1.4
-Summary: Lattifai Python SDK: Seamless Integration with Lattifai's Speech and Video AI Services
-Author-email: Lattifai Technologies <tech@lattifai.com>
-Maintainer-email: Lattice <tech@lattifai.com>
-License: MIT License
-        
-        Copyright (c) 2025 Lattifai.
-        
-        Permission is hereby granted, free of charge, to any person obtaining a copy
-        of this software and associated documentation files (the "Software"), to deal
-        in the Software without restriction, including without limitation the rights
-        to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
-        copies of the Software, and to permit persons to whom the Software is
-        furnished to do so, subject to the following conditions:
-        
-        The above copyright notice and this permission notice shall be included in all
-        copies or substantial portions of the Software.
-        
-        THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-        IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-        FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
-        AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
-        LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
-        OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
-        SOFTWARE.
-Project-URL: Homepage, https://github.com/lattifai/lattifai-python
-Project-URL: Documentation, https://github.com/lattifai/lattifai-python/api.md
-Project-URL: Bug Tracker, https://github.com/lattifai/lattifai-python/issues
-Project-URL: Discussions, https://github.com/lattifai/lattifai-python/discussions
-Project-URL: Changelog, https://github.com/lattifai/lattifai-python/CHANGELOG
-Keywords: lattifai,speech recognition,video analysis,ai,sdk,api client
-Classifier: Development Status :: 5 - Production/Stable
-Classifier: Intended Audience :: Developers
-Classifier: Intended Audience :: Science/Research
-Classifier: License :: OSI Approved :: Apache Software License
-Classifier: Programming Language :: Python :: 3.9
-Classifier: Programming Language :: Python :: 3.10
-Classifier: Programming Language :: Python :: 3.11
-Classifier: Programming Language :: Python :: 3.12
-Classifier: Programming Language :: Python :: 3.13
-Classifier: Operating System :: MacOS :: MacOS X
-Classifier: Operating System :: POSIX :: Linux
-Classifier: Operating System :: Microsoft :: Windows
-Classifier: Topic :: Multimedia :: Sound/Audio
-Classifier: Topic :: Multimedia :: Video
-Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
-Requires-Python: >=3.9
-Description-Content-Type: text/markdown
-License-File: LICENSE
-Requires-Dist: httpx
-Requires-Dist: python-dotenv
-Requires-Dist: lhotse>=1.26.0
-Requires-Dist: colorful>=0.5.6
-Requires-Dist: lattifai-core>=0.1.4
-Provides-Extra: numpy
-Requires-Dist: numpy; extra == "numpy"
-Provides-Extra: test
-Requires-Dist: pytest; extra == "test"
-Requires-Dist: pytest-cov; extra == "test"
-Requires-Dist: ruff; extra == "test"
-Requires-Dist: numpy; extra == "test"
-Provides-Extra: all
-Requires-Dist: numpy; extra == "all"
-Requires-Dist: pytest; extra == "all"
-Requires-Dist: pytest-cov; extra == "all"
-Requires-Dist: ruff; extra == "all"
-Dynamic: license-file
-
-# LattifAI Python
-
-> ⚠️ **Under Active Development** - Official release scheduled for October 18, 2025
-
-[![PyPI version](https://badge.fury.io/py/lattifai.svg)](https://badge.fury.io/py/lattifai)
-
-The official Python library for the LattifAI API - Advanced forced alignment and subtitle generation powered by `Lattice-1-Alpha` model.
-
-## Installation
-
-```bash
-pip install lattifai
-```
-
-## GPU Support Status
-
-> **🚧 GPU Support Coming Soon**: CUDA/GPU acceleration is currently under active development. The current version supports CPU-only processing. GPU support will be available in an upcoming release, providing significant performance improvements for large audio files.
-
-## Quick Start
-
-### Python API
-
-```python
-import os
-from lattifai import LattifAI
-
-# Initialize client
-client = LattifAI(
-    api_key=os.environ.get("LATTIFAI_API_KEY"),  # Optional if set in environment
-    device='cpu',  # Currently only CPU is supported, GPU support coming soon
-)
-
-# Perform alignment
-result = client.alignment(
-    audio="path/to/audio.wav",
-    subtitle="path/to/subtitle.srt",  # or .txt, .vtt, .ass
-    format="srt",  # auto-detect if not specified
-    output_subtitle_path="output.srt"
-)
-
-print(f"Alignment complete! Output saved to: {result}")
-```
-
-### Command Line Interface
-
-The library provides powerful command-line tools for batch processing and automation.
-
-#### Basic Alignment
-
-```bash
-# Align audio with subtitle file
-lattifai-align input_audio.wav input_subtitle.srt output_aligned.srt
-
-# Specify input format explicitly
-lattifai-align -F srt input_audio.wav transcript.txt output.srt
-
-# Auto-detect input format (default)
-lattifai-align input_audio.wav subtitle_file.vtt output.vtt
-```
-
-#### Using the Main CLI
-
-```bash
-# Show available commands
-lattifai --help
-
-# Perform alignment (same as lattifai-align)
-lattifai align input_audio.wav input_subtitle.srt output.srt
-
-# Subtitle format conversion
-lattifai subtitle convert input.srt output.vtt
-```
-
-#### Supported Input Formats
-
-- **SRT**: SubRip Subtitle format (`.srt`)
-- **VTT**: WebVTT format (`.vtt`)
-- **ASS**: Advanced SubStation Alpha format (`.ass`)
-- **TXT**: Plain text format (`.txt`)
-- **AUTO**: Automatic format detection (default)
-
-#### Command Line Options
-
-- `-F, --input_format`: Specify input subtitle format (`srt`, `vtt`, `ass`, `txt`, `auto`)
-- `--help`: Show help message and available options
-
-## API Reference
-
-### LattifAI Class
-
-#### Constructor Parameters
-
-```python
-LattifAI(
-    api_key: Optional[str] = None,           # API key (or set LATTIFAI_API_KEY env var)
-    base_url: Optional[str] = None,          # API base URL (or set LATTIFAI_BASE_URL env var)
-    device: str = 'cpu',                     # Device for processing (currently only 'cpu' supported)
-)
-```
-
-> **Note**: The `device` parameter currently only supports `'cpu'`. GPU support (`'cuda'`) is under active development and will be available in future releases.
-
-#### Methods
-
-##### `alignment()`
-
-Perform forced alignment between audio and text/subtitles.
-
-```python
-def alignment(
-    audio: Pathlike,                         # Path to audio file
-    subtitle: Pathlike,                      # Path to subtitle/text file
-    format: Optional[SubtitleFormat] = None, # Input format (auto-detect if None)
-    output_subtitle_path: Optional[Pathlike] = None  # Output file path
-) -> str:
-```
-
-**Parameters:**
-- `audio`: Path to the audio file (WAV, MP3, etc.)
-- `subtitle`: Path to subtitle or text file
-- `format`: Input format (`'srt'`, `'vtt'`, `'ass'`, `'txt'`, or `None` for auto-detection)
-- `output_subtitle_path`: Path where aligned subtitle will be saved
-
-**Returns:**
-- Path to output file (if `output_subtitle_path` specified) or alignment results
-
-**Example:**
-```python
-# Basic usage
-result = client.alignment("audio.wav", "subtitle.srt")
-
-# With explicit format and output path
-result = client.alignment(
-    audio="interview.mp3",
-    subtitle="transcript.txt",
-    format="txt",
-    output_subtitle_path="aligned_interview.srt"
-)
-```
-
-
-## Configuration
-
-### Environment Variables
-
-- `LATTIFAI_API_KEY`: Your LattifAI API key (required)
-- `LATTIFAI_BASE_URL`: Base URL for the API (default: `https://api.lattifai.com/v1`)
-
-### Device Configuration
-
-> **Note**: GPU/CUDA support is currently under development and will be available in a future release. Currently, only CPU processing is supported.
-
-The library is designed to support both CPU and GPU processing:
-
-```python
-# CPU processing (currently supported)
-client = LattifAI(device='cpu')
-
-# GPU processing (coming soon)
-# client = LattifAI(device='cuda')  # Will be available in future releases
-```
-
-### Performance Tuning
-
-For better performance with large files:
-
-```python
-client = LattifAI(
-    device='cpu',         # Currently only CPU is supported
-)
-```
-
-> **GPU Acceleration**: CUDA support is in active development and will significantly improve processing speed for large audio files. Expected in upcoming releases.
-
-## Examples
-
-### Basic Text Alignment
-
-```python
-from lattifai import LattifAI
-
-client = LattifAI()
-
-# Align plain text with audio
-result = client.alignment(
-    audio="speech.wav",
-    subtitle="Hello world. This is a test.",
-    format="txt",
-    output_subtitle_path="aligned.srt"
-)
-```
-
-### Subtitle Refinement
-
-```python
-# Improve existing subtitle timing
-result = client.alignment(
-    audio="movie.mp4",
-    subtitle="rough_subtitles.srt",
-    output_subtitle_path="precise_subtitles.srt"
-)
-```
-
-### Batch Processing
-
-```python
-import os
-from pathlib import Path
-
-client = LattifAI()
-
-# Process multiple files
-audio_dir = Path("audio_files")
-subtitle_dir = Path("subtitles")
-output_dir = Path("aligned_output")
-
-for audio_file in audio_dir.glob("*.wav"):
-    subtitle_file = subtitle_dir / f"{audio_file.stem}.srt"
-    output_file = output_dir / f"{audio_file.stem}_aligned.srt"
-
-    if subtitle_file.exists():
-        client.alignment(
-            audio=audio_file,
-            subtitle=subtitle_file,
-            output_subtitle_path=output_file
-        )
-```
-
-## Supported Formats
-
-### Audio Formats
-- **WAV**: Recommended for best quality
-- **MP3**: Widely supported compressed format
-- **FLAC**: Lossless compression
-- **M4A**: Apple audio format
-- **OGG**: Open source audio format
-
-### Subtitle Formats
-- **SRT**: SubRip Subtitle format (most common)
-- **VTT**: WebVTT format (web standard)
-- **ASS**: Advanced SubStation Alpha format (advanced styling)
-- **TXT**: Plain text format (one sentence per line)
-
-## Error Handling
-
-```python
-from lattifai import LattifAI, LattifAIError
-
-try:
-    client = LattifAI()
-    result = client.alignment(
-        audio="audio.wav",
-        subtitle="subtitle.srt"
-    )
-except LattifAIError as e:
-    print(f"LattifAI error: {e}")
-except FileNotFoundError as e:
-    print(f"File not found: {e}")
-except Exception as e:
-    print(f"Unexpected error: {e}")
-```
-
-### Common Error Scenarios
-
-- **Missing API Key**: Set `LATTIFAI_API_KEY` environment variable
-- **File Not Found**: Verify input file paths exist
-- **Unsupported Format**: Check supported audio/subtitle formats
-- **Network Issues**: Check internet connection and API endpoint
-- **Device Configuration**: Currently only `device='cpu'` is supported (GPU support coming soon)
-
-## Model Information
-
-This library uses the **Lattice-1-Alpha** model for high-quality forced alignment and subtitle generation.
-
-### Model Features
-- **High Accuracy**: State-of-the-art alignment precision
-- **Multi-language Support**: Supports various languages and accents
-- **Robust Processing**: Handles noisy audio and imperfect transcripts
-- **Fast Processing**: Optimized for both CPU and GPU execution
-
-### Model Requirements
-- **Minimum Python**: 3.9+
-- **Memory**: 4GB RAM recommended
-- **Storage**: ~2GB for model files
-- **GPU**: CUDA support coming soon (currently CPU-only)
-
-## Requirements
-
-- **Python**: 3.9-3.13
-- **Core Dependencies**:
-  - httpx (HTTP client)
-  - lhotse (audio processing)
-  - colorful (colored output)
-  - python-dotenv (environment management)
-  - click (command line interface)
-
-## Development
-
-### Setting Up Development Environment
-
-1. **Clone the repository:**
-```bash
-git clone https://github.com/lattifai/lattifai-python.git
-cd lattifai-python
-```
-
-2. **Install development dependencies:**
-```bash
-pip install -e ".[test]"
-```
-
-3. **Install Git hooks (recommended):**
-```bash
-./scripts/install-hooks.sh
-```
-
-This will install pre-commit hooks that automatically run `isort` and `ruff` checks before each commit to ensure code quality.
-
-### Code Quality Tools
-
-We use the following tools to maintain code quality:
-
-- **isort**: Automatically sorts and organizes imports
-- **ruff**: Fast Python linter and formatter
-
-To manually run these tools:
-
-```bash
-# Sort imports
-isort src/ tests/
-
-# Run linter
-ruff check src/ tests/
-
-# Run formatter
-ruff format src/ tests/
-
-# Fix issues automatically
-ruff check --fix src/ tests/
-```
-
-### Running Tests
-
-```bash
-# Run all tests
-pytest
-
-# Run with coverage
-pytest --cov=src --cov-report=html
-
-# Run specific test file
-pytest tests/test_basic.py -v
-```
-
-### Building and Testing Locally
-
-```bash
-# Build package
-python -m build
-
-# Install locally
-pip install -e .
-
-# Test CLI commands
-lattifai --help
-lattifai-align --help
-```
-
-### Contributing
-
-1. Fork the repository
-2. Create a feature branch: `git checkout -b feature-name`
-3. Make your changes and add tests
-4. Run the test suite: `pytest`
-5. Run code quality checks: `ruff check src/ tests/`
-6. Commit your changes: `git commit -am 'Add some feature'`
-7. Push to the branch: `git push origin feature-name`
-8. Submit a pull request
-
-### Bypassing Pre-commit Hooks
-
-If you need to commit without running the hooks (not recommended):
-
-```bash
-git commit --no-verify
-```
-
-## License
-
-This project is licensed under the Apache License 2.0 - see the [LICENSE](LICENSE) file for details.
-
-## Support
-
-- **Documentation**: [API Documentation](https://github.com/lattifai/lattifai-python/api.md)
-- **Issues**: [GitHub Issues](https://github.com/lattifai/lattifai-python/issues)
-- **Discussions**: [GitHub Discussions](https://github.com/lattifai/lattifai-python/discussions)
-- **Changelog**: [CHANGELOG.md](https://github.com/lattifai/lattifai-python/CHANGELOG)

lattifai-0.1.4.dist-info/RECORD

@@ -1,22 +0,0 @@
-lattifai/__init__.py,sha256=8vZF9_yObaoDZ6sUaJBuQqUp5mGBRfD4z0SdTY_8BcQ,672
-lattifai/base_client.py,sha256=ktFtATjL9pLSJUD-VqeJKA1FHkrsGHX7Uq_x00H7gO8,3322
-lattifai/client.py,sha256=szz9aT_GTa5JKA2ZGsd7XHr56aG_H-nf83yqRcPUVxs,4664
-lattifai/bin/__init__.py,sha256=7YhmtEM8kbxJtz2-KIskvpLKBZAvkMSceVx8z4fkgQ4,61
-lattifai/bin/align.py,sha256=uX8VaATzn8CgdHUtry1ZGhXiz0Jr89ELdfRK6GWC1t8,989
-lattifai/bin/cli_base.py,sha256=4xlN4cnJZh54ErhHUHgJpyVsrcB-ftwniFzRQL_7SlU,289
-lattifai/bin/subtitle.py,sha256=bUWImAHpvyY59Vskqb5loQiD5ytQOxR8lTQRiQ4LyNA,647
-lattifai/io/__init__.py,sha256=vHWRN7MvAch-GUeFqqO-gM57SM-4YOpGUjIxFJdjfPA,671
-lattifai/io/reader.py,sha256=ErPnPMUvYQpjZ7Vd86EsHUkOcEfKdoI8iM3yKHRzSOQ,2576
-lattifai/io/supervision.py,sha256=5UfSsgBhXoDU3-6drDtoD7y8HIiA4xRKZnbOKgeejwM,354
-lattifai/io/writer.py,sha256=1eAEFLlL8kricxRDPFBtVmeC4IiFyFnjbWXvw0VU-q4,2036
-lattifai/tokenizers/__init__.py,sha256=aqv44PDtq6g3oFFKW_l4HSR5ywT5W8eP1dHHywIvBfs,72
-lattifai/tokenizers/phonemizer.py,sha256=Q5Z-4rbT3AjAPLNPnyvWGcEaJuKXRudgeIK6tUhVsJs,1741
-lattifai/tokenizers/tokenizer.py,sha256=Qqg12zihl192Tlax6plVyxthrnzBciGLSRuzFQRgOdc,5663
-lattifai/workers/__init__.py,sha256=s6YfkIq4FDIAzY9sPjRpXnJfszj2repqnMTqydRM5Zw,83
-lattifai/workers/lattice1_alpha.py,sha256=kR5wNLMn1qN14PvRA6RlWjQUGblYrJ636ILC-XkvS0s,4770
-lattifai-0.1.4.dist-info/licenses/LICENSE,sha256=LNuoH5jpXXNKgjQ3XLwztFq8D3O7kZI-LSg81o4ym2M,1065
-lattifai-0.1.4.dist-info/METADATA,sha256=HEsoKRRfjRoeNwUff7Cx44sYP9trid_bdtX3j5WqK6o,13628
-lattifai-0.1.4.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-lattifai-0.1.4.dist-info/entry_points.txt,sha256=BaixG8HYzE3Ff0QDAmsbX8cjsxZSLwUv5u4cnIpunr0,88
-lattifai-0.1.4.dist-info/top_level.txt,sha256=tHSoXF26r-IGfbIP_JoYATqbmf14h5NrnNJGH4j5reI,9
-lattifai-0.1.4.dist-info/RECORD,,