kgnlp/allophant
A multilingual phoneme recognizer capable of generalizing zero-shot to unseen phoneme inventories.
This tool helps researchers and linguists automatically identify individual speech sounds (phonemes) from spoken audio, even for languages it hasn't specifically been trained on. You provide an audio file, and it outputs a sequence of recognized phonemes or phonetic features for that speech. It's designed for speech scientists, phoneticians, and language technology developers working with diverse linguistic data.
No commits in the last 6 months. Available on PyPI.
Use this if you need to extract phoneme sequences or phonetic features from spoken audio in various languages, including those with unique sound systems that are typically hard for existing tools to handle.
Not ideal if you need a simple, off-the-shelf speech-to-text transcriber for common languages, or if you don't have experience working with phonetic inventories and Python.
Stars
29
Forks
3
Language
Python
License
MIT
Category
Last pushed
Mar 14, 2025
Commits (30d)
0
Dependencies
24
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/kgnlp/allophant"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
espnet/espnet
End-to-End Speech Processing Toolkit
yeyupiaoling/PPASR
基于PaddlePaddle实现端到端中文语音识别，从入门到实战，超简单的入门案例，超实用的企业项目。支持当前最流行的DeepSpeech2、Conformer、Squeezeformer模型
yeyupiaoling/PaddlePaddle-DeepSpeech
基于PaddlePaddle实现的语音识别，中文语音识别。项目完善，识别效果好。支持Windows，Linux下训练和预测，支持Nvidia Jetson开发板预测。
flashlight/wav2letter
Facebook AI Research's Automatic Speech Recognition Toolkit
pannous/tensorflow-speech-recognition
🎙Speech recognition using the tensorflow deep learning framework, sequence-to-sequence neural networks