Unified model for code-switching speech recognition and language identification based on a concatenated tokenizer

Open in new window