# mravanelli/SincNet

SincNet is a neural architecture for efficiently processing raw audio samples.

Repository: https://github.com/mravanelli/SincNet
Canonical: https://ross.abutalabs.com/products/sincnet
Language: Python
License: MIT
License Family: permissive
Topics: deep-learning, audio, waveform, filtering, cnn, convolutional-neural-networks, speaker-recognition, speaker-verification, speaker-identification, speech-recognition, asr, audio-processing, speech-processing, digital-signal-processing, signal-processing, neural-networks, artificial-intelligence, timit, pytorch, python
Last push: 2021-04-28T07:54:52+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2976, "days_push": 1953, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1243, forks 272 (observed 2026-08-28T04:04:06.692090+00:00)

## What it is
SincNet is a PyTorch neural architecture that processes raw audio waveforms using parametrized sinc band-pass filters in the first convolutional layer. The repository provides code and utilities for speaker identification, with a TIMIT example included.

## Use cases
- speaker identification from raw audio waveforms
- speaker verification with deep learning
- learn customized filter banks for audio tasks
- run speaker recognition experiments on TIMIT
- process raw audio samples with CNNs instead of spectrograms
- research learnable front-end filters for speech models

## When to choose
- you want to learn filters directly from raw waveforms rather than hand-crafted spectrogram features
- you need a compact parametrized band-pass filter bank tuned to your audio application
- you are doing speaker recognition research and want a well-known reference implementation

## When to avoid
- you need a maintained general-purpose speech toolkit - consider SpeechBrain or PyTorch-Kaldi instead
- you want production-ready ASR rather than a research showcase
- you work outside Linux or need recent PyTorch version support

## Facets
- artifact type: library
- maturity: maintenance
- function: audio-processing, machine-learning, deep-learning, speech-recognition
- domain: speech-processing, deep-learning, machine-learning, artificial-intelligence
- platform: python
- tags: sinc-filters, raw-waveform, speaker-recognition, cnn, pytorch, timit, digital-signal-processing, speaker-verification, audio, linux, gpu

## Member repositories
- mravanelli/SincNet (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:06.692090+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T05:08:20.243760+00:00, confidence not recorded.
  - readme: https://github.com/mravanelli/SincNet (fetched 2026-08-28T04:04:06.692090+00:00, sha 7883d0251b94)
- Data as of 2026-08-30T08:39:29.467469+00:00.