pydeepspeech
- The simpliest way to use AI to generate transcriptions from a wav file.
- This project uses the Mozilla DeepSpeech engine built from the included demo:
Why you need this
Mozilla's deep speech can't process long voice samples. pydeepspeech fixes this by "chunking" the input sound into seperate wav files that are then individualy processed. Wav files are cut along periods of detected silence, controlled by the aggressive parameter.
Besides this, pydeepspeech is probably better to use anyway because it's much simpler to install than Mozilla's Deepspeech because the required data models needed for pydeepspeech are automatically downloaded and installed on first use.
Quick start
Console api:
$ pip install pydeepspeech
$ pydeepspeech --wav_file <WAVE_FILE> --aggressive 1 --out_file <TEXT_FILE>
-or-
$ pip install pydeepspeech
$ pydeepspeech --wav_file <WAVE_FILE> --out_file <TEXT_FILE> --model_dir <MY_PBMM_AND_SCORER_FILES>
-or-
$ pip install pydeepspeech
$ pydeepspeech_installmodels --pbmm <PBMM_FILE_OR_URL> --scorer <SCORER_FILE_OR_URL>
$ pydeepspeech --wav_file <WAVE_FILE> --out_file <TEXT_FILE>
Or in python
from pydeepspeech.transcribe import transcribe
transcribe(...)
Optional: Create a virtual python package
Download and install virtual env:
# Download
curl -X GET https://raw.githubusercontent.com/zackees/make_venv/main/make_venv.py -o make_env.py
python make_env.py # Make the environment
source activate.sh # Enter environment
$ pip install pydeepspeech
To get back into the environment execute source activate.sh (if windows, you must be using git-bash)
Testing
Testing and linting is very simple. Just run tox (link).
$ pip install tox
$ tox
Metadata
Release files for pydeepspeech 1.1.7
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| pydeepspeech-1.1.7.tar.gz | 11.1 kB | Without provenance | Details for pydeepspeech-1.1.7.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| pydeepspeech-1.1.7-py2.py3-none-any.whl | Python 2, Python 3 | none | any | Details for pydeepspeech-1.1.7-py2.py3-none-any.whl |
Total release size: 26.4 kB
Release files / pydeepspeech-1.1.7.tar.gz
| Download URL | pydeepspeech-1.1.7.tar.gz |
|---|---|
| Size | 11.1 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
1b56517d032791714358e9d73a926886e22f748202ea67798b9f4fccf332be2c
|
|
BLAKE2b-256 checksum How to use checksums |
f15476c4a7880a27b83f38573c4c75c9c636d41d4e25785ff5e3d655677ad0c4
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/3.7.1 importlib_metadata/4.9.0 pkginfo/1.8.2 requests/2.26.0 requests-toolbelt/0.9.1 tqdm/4.62.3 CPython/3.8.0
|
Release files / pydeepspeech-1.1.7-py2.py3-none-any.whl
| Download URL | pydeepspeech-1.1.7-py2.py3-none-any.whl |
|---|---|
| Size | 15.3 kB |
| Tags | Python 2 Python 3 |
|
SHA-256 checksum How to use checksums |
22597a259cf4faef6e9104e3780ba8923e5b03debfed46c52956f7f95336cc59
|
|
BLAKE2b-256 checksum How to use checksums |
7dff8dfb90c3b4c87d78deefd4530605393d7d2fb159ffa84f51adfca4cb783b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/3.7.1 importlib_metadata/4.9.0 pkginfo/1.8.2 requests/2.26.0 requests-toolbelt/0.9.1 tqdm/4.62.3 CPython/3.8.0
|
Security for pydeepspeech 1.1.7
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.