Skip to main content

HOJO ASR inference library

Project description

hojo-asr

HOJO ASR inference library

The bundled WeNet-derived code lives under hojo_asr.wenet (not the top-level wenet package) to avoid import conflicts with the official WeNet project.

Installation

pip install hojo-asr

Install from source for development:

cd inference/hojo_asr
pip install -e .

Quick start

from hojo_asr import HOJO_ASR

model = HOJO_ASR.load_model("/path/to/model_folder", device="cuda:0")
results = model.run_infer(["/path/to/audio.wav"], batch_size=1)

for item in results:
    print(item["key"], item["text"])

The model folder should contain:

  • merged_full_model.safetensors
  • config.yaml
  • Subdirectories referenced in config.yaml

Publish to PyPI

cd inference/hojo_asr
python -m pip install build twine
python -m build
twine upload dist/*

Third-party notices

This package includes a trimmed subset of WeNet (Apache License 2.0). Some modules are also derived from ESPnet (Apache License 2.0).

See NOTICE for attribution and a summary of modifications to the bundled code under hojo_asr.wenet. This package is not the official WeNet distribution and does not install a top-level wenet module.

License

Copyright 2026 HojoAI.

Licensed under the Apache License, Version 2.0. See LICENSE and NOTICE for details.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

hojo_asr-0.1.0.tar.gz (49.4 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

hojo_asr-0.1.0-py3-none-any.whl (64.8 kB view details)

Uploaded Python 3

File details

Details for the file hojo_asr-0.1.0.tar.gz.

File metadata

  • Download URL: hojo_asr-0.1.0.tar.gz
  • Upload date:
  • Size: 49.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.10.19

File hashes

Hashes for hojo_asr-0.1.0.tar.gz
Algorithm Hash digest
SHA256 a7741ef95b5a47fdbf387eea0e1758682b7f76465d42110bf2e739677551c59f
MD5 d8b99f8945e556d5231df51be23ae7c0
BLAKE2b-256 fe626cd29cc8712100e28a9cb9157e3e80227056f9750747015776081c6a154c

See more details on using hashes here.

File details

Details for the file hojo_asr-0.1.0-py3-none-any.whl.

File metadata

  • Download URL: hojo_asr-0.1.0-py3-none-any.whl
  • Upload date:
  • Size: 64.8 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.10.19

File hashes

Hashes for hojo_asr-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 f95a5bdd9aee684fd5240efadaae0bd629ae4e3949755ff6026ee1d370953cb7
MD5 89718b531ee0a541d9959853936d671f
BLAKE2b-256 a3e38c92dea4f49319cdef8fece18f710d7ae2335cbafb984f749ee08dcd1f1c

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page