paratran 0.1.0__tar.gz

paratran-0.1.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Brian Sunter
+
+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.

paratran-0.1.0/PKG-INFO

@@ -0,0 +1,134 @@
+Metadata-Version: 2.4
+Name: paratran
+Version: 0.1.0
+Summary: REST API for audio transcription using parakeet-mlx
+Author: Brian Sunter
+License-Expression: MIT
+Project-URL: Homepage, https://github.com/briansunter/paratran
+Project-URL: Repository, https://github.com/briansunter/paratran
+Keywords: transcription,asr,speech-to-text,mlx,apple-silicon
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Operating System :: MacOS
+Classifier: Programming Language :: Python :: 3
+Classifier: Topic :: Multimedia :: Sound/Audio :: Speech
+Requires-Python: >=3.11
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: parakeet-mlx
+Requires-Dist: fastapi
+Requires-Dist: uvicorn[standard]
+Requires-Dist: python-multipart
+Dynamic: license-file
+
+# Paratran
+
+REST API for audio transcription on Apple Silicon, powered by [parakeet-mlx](https://github.com/senstella/parakeet-mlx).
+
+Parakeet is #1 on the [Open ASR Leaderboard](https://huggingface.co/spaces/hf-audio/open_asr_leaderboard) and runs ~30x faster than Whisper on Apple Silicon via MLX.
+
+## Requirements
+
+- macOS with Apple Silicon (M1/M2/M3/M4)
+- Python 3.11+
+- ~2 GB memory for the default model
+
+## Quick Start
+
+```bash
+git clone https://github.com/briansunter/paratran.git
+cd paratran
+./run.sh
+```
+
+This creates a virtual environment, installs dependencies, downloads the model, and starts the server at `http://localhost:8000`.
+
+## Install
+
+```bash
+pip install paratran
+paratran
+```
+
+Or from source:
+
+```bash
+python3 -m venv .venv
+source .venv/bin/activate
+pip install -e .
+paratran
+```
+
+## Usage
+
+```bash
+# Default settings
+paratran
+
+# Custom model cache location
+paratran --model-dir /Volumes/Storage/models
+
+# Custom host and port
+paratran --host 127.0.0.1 --port 9000
+
+# Different model
+paratran --model mlx-community/parakeet-tdt-1.1b-v2
+```
+
+All options can also be set via environment variables:
+
+| CLI Flag       | Env Var              | Default                                  |
+|----------------|----------------------|------------------------------------------|
+| `--model`      | `PARATRAN_MODEL`     | `mlx-community/parakeet-tdt-0.6b-v3`    |
+| `--model-dir`  | `PARATRAN_MODEL_DIR` | HuggingFace default (`~/.cache/huggingface`) |
+| `--host`       |                      | `0.0.0.0`                                |
+| `--port`       |                      | `8000`                                   |
+
+## API
+
+### `GET /health`
+
+```bash
+curl http://localhost:8000/health
+```
+
+```json
+{
+  "status": "ok",
+  "model": "mlx-community/parakeet-tdt-0.6b-v3",
+  "model_dir": "/Volumes/Storage/models"
+}
+```
+
+### `POST /transcribe`
+
+Upload an audio file (wav, mp3, flac, m4a, ogg, webm):
+
+```bash
+curl -X POST http://localhost:8000/transcribe -F "file=@recording.m4a"
+```
+
+```json
+{
+  "text": "Hello world, this is a test.",
+  "duration": 3.52,
+  "processing_time": 0.176,
+  "sentences": [
+    {
+      "text": "Hello world, this is a test.",
+      "start": 0.0,
+      "end": 3.52,
+      "tokens": [
+        { "text": "Hello", "start": 0.0, "end": 0.48 },
+        { "text": " world", "start": 0.48, "end": 0.8 }
+      ]
+    }
+  ]
+}
+```
+
+Interactive API docs are available at `http://localhost:8000/docs`.
+
+## License
+
+MIT

paratran-0.1.0/README.md

@@ -0,0 +1,111 @@
+# Paratran
+
+REST API for audio transcription on Apple Silicon, powered by [parakeet-mlx](https://github.com/senstella/parakeet-mlx).
+
+Parakeet is #1 on the [Open ASR Leaderboard](https://huggingface.co/spaces/hf-audio/open_asr_leaderboard) and runs ~30x faster than Whisper on Apple Silicon via MLX.
+
+## Requirements
+
+- macOS with Apple Silicon (M1/M2/M3/M4)
+- Python 3.11+
+- ~2 GB memory for the default model
+
+## Quick Start
+
+```bash
+git clone https://github.com/briansunter/paratran.git
+cd paratran
+./run.sh
+```
+
+This creates a virtual environment, installs dependencies, downloads the model, and starts the server at `http://localhost:8000`.
+
+## Install
+
+```bash
+pip install paratran
+paratran
+```
+
+Or from source:
+
+```bash
+python3 -m venv .venv
+source .venv/bin/activate
+pip install -e .
+paratran
+```
+
+## Usage
+
+```bash
+# Default settings
+paratran
+
+# Custom model cache location
+paratran --model-dir /Volumes/Storage/models
+
+# Custom host and port
+paratran --host 127.0.0.1 --port 9000
+
+# Different model
+paratran --model mlx-community/parakeet-tdt-1.1b-v2
+```
+
+All options can also be set via environment variables:
+
+| CLI Flag       | Env Var              | Default                                  |
+|----------------|----------------------|------------------------------------------|
+| `--model`      | `PARATRAN_MODEL`     | `mlx-community/parakeet-tdt-0.6b-v3`    |
+| `--model-dir`  | `PARATRAN_MODEL_DIR` | HuggingFace default (`~/.cache/huggingface`) |
+| `--host`       |                      | `0.0.0.0`                                |
+| `--port`       |                      | `8000`                                   |
+
+## API
+
+### `GET /health`
+
+```bash
+curl http://localhost:8000/health
+```
+
+```json
+{
+  "status": "ok",
+  "model": "mlx-community/parakeet-tdt-0.6b-v3",
+  "model_dir": "/Volumes/Storage/models"
+}
+```
+
+### `POST /transcribe`
+
+Upload an audio file (wav, mp3, flac, m4a, ogg, webm):
+
+```bash
+curl -X POST http://localhost:8000/transcribe -F "file=@recording.m4a"
+```
+
+```json
+{
+  "text": "Hello world, this is a test.",
+  "duration": 3.52,
+  "processing_time": 0.176,
+  "sentences": [
+    {
+      "text": "Hello world, this is a test.",
+      "start": 0.0,
+      "end": 3.52,
+      "tokens": [
+        { "text": "Hello", "start": 0.0, "end": 0.48 },
+        { "text": " world", "start": 0.48, "end": 0.8 }
+      ]
+    }
+  ]
+}
+```
+
+Interactive API docs are available at `http://localhost:8000/docs`.
+
+## License
+
+MIT

paratran-0.1.0/paratran/__init__.py

@@ -0,0 +1,3 @@
+"""Paratran — audio transcription REST API powered by parakeet-mlx."""
+
+__version__ = "0.1.0"

paratran-0.1.0/paratran/cli.py

@@ -0,0 +1,20 @@
+import argparse
+
+from paratran.server import DEFAULT_MODEL, MODEL_DIR, MODEL_NAME
+
+
+def main():
+    from paratran import server
+
+    parser = argparse.ArgumentParser(description="Paratran transcription server")
+    parser.add_argument("--model", default=MODEL_NAME, help=f"HF model ID or local path (default: {DEFAULT_MODEL})")
+    parser.add_argument("--model-dir", default=MODEL_DIR, help="Directory to download/cache models")
+    parser.add_argument("--host", default="0.0.0.0", help="Bind host (default: 0.0.0.0)")
+    parser.add_argument("--port", type=int, default=8000, help="Bind port (default: 8000)")
+    args = parser.parse_args()
+
+    server.MODEL_NAME = args.model
+    server.MODEL_DIR = args.model_dir
+
+    import uvicorn
+    uvicorn.run(server.app, host=args.host, port=args.port)

paratran-0.1.0/paratran/server.py

@@ -0,0 +1,77 @@
+import os
+import tempfile
+import time
+from pathlib import Path
+
+import parakeet_mlx
+from fastapi import FastAPI, File, UploadFile
+from fastapi.responses import JSONResponse
+
+DEFAULT_MODEL = "mlx-community/parakeet-tdt-0.6b-v3"
+MODEL_NAME = os.environ.get("PARATRAN_MODEL", DEFAULT_MODEL)
+MODEL_DIR = os.environ.get("PARATRAN_MODEL_DIR", None)
+ALLOWED_EXTENSIONS = {".wav", ".mp3", ".flac", ".m4a", ".ogg", ".webm"}
+
+app = FastAPI(title="Paratran", description="Audio transcription API powered by parakeet-mlx")
+model = None
+
+
+@app.on_event("startup")
+def load_model():
+    global model
+    kwargs = {}
+    if MODEL_DIR:
+        kwargs["cache_dir"] = MODEL_DIR
+    model = parakeet_mlx.from_pretrained(MODEL_NAME, **kwargs)
+
+
+@app.get("/health")
+def health():
+    return {"status": "ok", "model": MODEL_NAME, "model_dir": MODEL_DIR}
+
+
+@app.post("/transcribe")
+async def transcribe(
+    file: UploadFile = File(...),
+):
+    suffix = Path(file.filename).suffix.lower() if file.filename else ""
+    if suffix not in ALLOWED_EXTENSIONS:
+        return JSONResponse(
+            status_code=400,
+            content={"error": f"Unsupported file type '{suffix}'. Allowed: {', '.join(sorted(ALLOWED_EXTENSIONS))}"},
+        )
+
+    tmp = None
+    try:
+        tmp = tempfile.NamedTemporaryFile(suffix=suffix, delete=False)
+        tmp.write(await file.read())
+        tmp.close()
+
+        start = time.perf_counter()
+        result = model.transcribe(tmp.name)
+        elapsed = time.perf_counter() - start
+
+        sentences = []
+        for seg in result.sentences:
+            tokens = [
+                {"text": t.text, "start": t.start, "end": t.end}
+                for t in seg.tokens
+            ]
+            sentences.append({
+                "text": seg.text,
+                "start": seg.start,
+                "end": seg.end,
+                "tokens": tokens,
+            })
+
+        duration = result.sentences[-1].end if result.sentences else 0.0
+
+        return {
+            "text": result.text,
+            "duration": duration,
+            "processing_time": round(elapsed, 3),
+            "sentences": sentences,
+        }
+    finally:
+        if tmp:
+            Path(tmp.name).unlink(missing_ok=True)

paratran-0.1.0/paratran.egg-info/PKG-INFO

@@ -0,0 +1,134 @@
+Metadata-Version: 2.4
+Name: paratran
+Version: 0.1.0
+Summary: REST API for audio transcription using parakeet-mlx
+Author: Brian Sunter
+License-Expression: MIT
+Project-URL: Homepage, https://github.com/briansunter/paratran
+Project-URL: Repository, https://github.com/briansunter/paratran
+Keywords: transcription,asr,speech-to-text,mlx,apple-silicon
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Operating System :: MacOS
+Classifier: Programming Language :: Python :: 3
+Classifier: Topic :: Multimedia :: Sound/Audio :: Speech
+Requires-Python: >=3.11
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: parakeet-mlx
+Requires-Dist: fastapi
+Requires-Dist: uvicorn[standard]
+Requires-Dist: python-multipart
+Dynamic: license-file
+
+# Paratran
+
+REST API for audio transcription on Apple Silicon, powered by [parakeet-mlx](https://github.com/senstella/parakeet-mlx).
+
+Parakeet is #1 on the [Open ASR Leaderboard](https://huggingface.co/spaces/hf-audio/open_asr_leaderboard) and runs ~30x faster than Whisper on Apple Silicon via MLX.
+
+## Requirements
+
+- macOS with Apple Silicon (M1/M2/M3/M4)
+- Python 3.11+
+- ~2 GB memory for the default model
+
+## Quick Start
+
+```bash
+git clone https://github.com/briansunter/paratran.git
+cd paratran
+./run.sh
+```
+
+This creates a virtual environment, installs dependencies, downloads the model, and starts the server at `http://localhost:8000`.
+
+## Install
+
+```bash
+pip install paratran
+paratran
+```
+
+Or from source:
+
+```bash
+python3 -m venv .venv
+source .venv/bin/activate
+pip install -e .
+paratran
+```
+
+## Usage
+
+```bash
+# Default settings
+paratran
+
+# Custom model cache location
+paratran --model-dir /Volumes/Storage/models
+
+# Custom host and port
+paratran --host 127.0.0.1 --port 9000
+
+# Different model
+paratran --model mlx-community/parakeet-tdt-1.1b-v2
+```
+
+All options can also be set via environment variables:
+
+| CLI Flag       | Env Var              | Default                                  |
+|----------------|----------------------|------------------------------------------|
+| `--model`      | `PARATRAN_MODEL`     | `mlx-community/parakeet-tdt-0.6b-v3`    |
+| `--model-dir`  | `PARATRAN_MODEL_DIR` | HuggingFace default (`~/.cache/huggingface`) |
+| `--host`       |                      | `0.0.0.0`                                |
+| `--port`       |                      | `8000`                                   |
+
+## API
+
+### `GET /health`
+
+```bash
+curl http://localhost:8000/health
+```
+
+```json
+{
+  "status": "ok",
+  "model": "mlx-community/parakeet-tdt-0.6b-v3",
+  "model_dir": "/Volumes/Storage/models"
+}
+```
+
+### `POST /transcribe`
+
+Upload an audio file (wav, mp3, flac, m4a, ogg, webm):
+
+```bash
+curl -X POST http://localhost:8000/transcribe -F "file=@recording.m4a"
+```
+
+```json
+{
+  "text": "Hello world, this is a test.",
+  "duration": 3.52,
+  "processing_time": 0.176,
+  "sentences": [
+    {
+      "text": "Hello world, this is a test.",
+      "start": 0.0,
+      "end": 3.52,
+      "tokens": [
+        { "text": "Hello", "start": 0.0, "end": 0.48 },
+        { "text": " world", "start": 0.48, "end": 0.8 }
+      ]
+    }
+  ]
+}
+```
+
+Interactive API docs are available at `http://localhost:8000/docs`.
+
+## License
+
+MIT

paratran-0.1.0/paratran.egg-info/SOURCES.txt

@@ -0,0 +1,12 @@
+LICENSE
+README.md
+pyproject.toml
+paratran/__init__.py
+paratran/cli.py
+paratran/server.py
+paratran.egg-info/PKG-INFO
+paratran.egg-info/SOURCES.txt
+paratran.egg-info/dependency_links.txt
+paratran.egg-info/entry_points.txt
+paratran.egg-info/requires.txt
+paratran.egg-info/top_level.txt

paratran-0.1.0/paratran.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

paratran-0.1.0/paratran.egg-info/entry_points.txt

@@ -0,0 +1,2 @@
+[console_scripts]
+paratran = paratran.cli:main

paratran-0.1.0/paratran.egg-info/requires.txt

@@ -0,0 +1,4 @@
+parakeet-mlx
+fastapi
+uvicorn[standard]
+python-multipart

paratran-0.1.0/paratran.egg-info/top_level.txt

@@ -0,0 +1 @@
+paratran

paratran-0.1.0/pyproject.toml

@@ -0,0 +1,33 @@
+[build-system]
+requires = ["setuptools>=68.0"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "paratran"
+version = "0.1.0"
+description = "REST API for audio transcription using parakeet-mlx"
+readme = "README.md"
+license = "MIT"
+requires-python = ">=3.11"
+authors = [{ name = "Brian Sunter" }]
+keywords = ["transcription", "asr", "speech-to-text", "mlx", "apple-silicon"]
+classifiers = [
+    "Development Status :: 3 - Alpha",
+    "Intended Audience :: Developers",
+    "Operating System :: MacOS",
+    "Programming Language :: Python :: 3",
+    "Topic :: Multimedia :: Sound/Audio :: Speech",
+]
+dependencies = [
+    "parakeet-mlx",
+    "fastapi",
+    "uvicorn[standard]",
+    "python-multipart",
+]
+
+[project.urls]
+Homepage = "https://github.com/briansunter/paratran"
+Repository = "https://github.com/briansunter/paratran"
+
+[project.scripts]
+paratran = "paratran.cli:main"

paratran-0.1.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+