File size: 2,388 Bytes
6546267 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 | ---
license: apache-2.0
language: fa
library_name: onnx
tags:
- onnx
- wav2vec2
- ctc
- forced-alignment
- quantized
- parseh
base_model: jonatasgrosman/wav2vec2-large-xlsr-53-persian
base_model_relation: quantized
datasets:
- common_voice
---
# Parseh Persian forced-alignment network
This CTC network aligns known Persian caption text to 16 kHz mono audio in Parseh. It is not a general-purpose speech-recognition model. Parseh runs it locally with NumPy and ONNX Runtime, then uses CTC Viterbi alignment for word spans or character spans.
## Provenance and conversion
Derived from [jonatasgrosman/wav2vec2-large-xlsr-53-persian](https://huggingface.co/jonatasgrosman/wav2vec2-large-xlsr-53-persian) at `234714078a1398a9db88194c5a40fefe6f376dc1`, based on `facebook/wav2vec2-large-xlsr-53`. It was converted to ONNX opset 17 and dynamically quantized to int8 (`QUInt8`, per-channel `MatMul` weights); no retraining was performed. Exact source and output hashes are in `meta.json` and `SHA256SUMS`.
## Verification
- ONNX Runtime CPU versions: 1.23.2, 1.30.0; dynamic 0.5 s and 60 s inputs passed.
- fp32 ONNX/PyTorch log-softmax max absolute difference: `0.000247955322265625`; frame-argmax agreement: `1.0`.
- int8/fp32 frame-argmax agreement: `1.0`; generated-speech span-start delta median/p95/max: `0.0` / `0.0` / `0.0` ms.
- Generated 30-second CPU alignment cost: `5.826939` s; peak RSS: `2907074560` bytes.
`espeak-ng` generated audio was used only for pipeline monotonicity, not a real-speech accuracy claim. Unknown characters use a wildcard score to preserve a monotonic path.
## Limitations
Input must be 16 kHz mono audio and known caption text. Noise, accents, text errors, vocabulary gaps, and quantization can affect timestamps. No accuracy guarantee is made.
## Licence and attribution
The source tag is `apache-2.0`. `LICENSE` and `NOTICE` retain source, base-model, dataset attribution, and changes. No endorsement by original authors, Meta, Mozilla, or Parseh is implied.
## Citation from the source card
```bibtex
@misc{grosman2021xlsr53-large-persian,
title={Fine-tuned {XLSR}-53 large model for speech recognition in {P}ersian},
author={Grosman, Jonatas},
howpublished={\url{https://huggingface.co/jonatasgrosman/wav2vec2-large-xlsr-53-persian}},
year={2021}
}
```
## Contact
Contact: [development@parseh.io](mailto:development@parseh.io).
|