Implementation:Speechbrain Speechbrain Prepare CommonVoice LM
| Knowledge Sources | |
|---|---|
| Domains | Language Modeling, Data Preparation |
| Last Updated | 2026-02-09 00:00 GMT |
Overview
Variant of the Common Voice data preparation script for the language model recipe.
Description
This file (recipes/CommonVoice/LM/common_voice_prepare.py) is a copy of the canonical Common Voice data preparation script. It provides the same prepare_common_voice function with identical parameters and behavior, placed within the LM recipe subdirectory for convenience. The canonical implementation is documented on the Implementation:Speechbrain_Speechbrain_Prepare_CommonVoice_Seq2Seq page.
Usage
Use this when preparing the Mozilla Common Voice dataset specifically for language model training. See the canonical page for full details.
Code Reference
Source Location
- Repository: SpeechBrain
- File: recipes/CommonVoice/LM/common_voice_prepare.py
Canonical Implementation
This is a duplicate of the seq2seq Common Voice preparation script. For full documentation including signature, I/O contract, and usage examples, see:
Related Pages
- Implementation:Speechbrain_Speechbrain_Prepare_CommonVoice_Seq2Seq -- Canonical implementation page
- Implementation:Speechbrain_Speechbrain_Prepare_CommonVoice_Transducer -- Same script used for transducer recipe
- Implementation:Speechbrain_Speechbrain_Prepare_CommonVoice_SSL -- Same script used for self-supervised learning recipe
- Principle:Speechbrain_Speechbrain_Dataset_Specific_Data_Preparation