ASR-project/Multilingual-PR
Phoneme Recognition using pre-trained models Wav2vec2, HuBERT and WavLM. Throughout this project, we compared specifically three different self-supervised models, Wav2vec (2019, 2020), HuBERT (2021) and WavLM (2022) pretrained on a corpus of English speech that we will use in various ways to perform phoneme recognition for different languages with a network trained with Connectionist Temporal Classification (CTC) algorithm.
This project helps speech scientists or linguists analyze spoken language by converting audio recordings into sequences of phonetic units (phonemes). It takes audio data in various languages and outputs a detailed transcription of the phonemes spoken. This tool is ideal for researchers studying multilingual speech patterns, phonetic transfer, or those developing language technologies.
258 stars. No commits in the last 6 months.
Use this if you need to accurately identify and transcribe phonemes from audio recordings across multiple languages, especially when working with limited annotated data.
Not ideal if you need full word-level transcription or speaker identification rather than just phoneme sequences.
Stars
258
Forks
22
Language
Python
License
—
Category
Last pushed
May 09, 2022
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/ASR-project/Multilingual-PR"
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