IBM 发布 Granite-Speech-5.0-2B 多语言语音模型
原标题：ibm-granite/granite-speech-5.0-2b
AI 摘要
IBM 在 Hugging Face 发布 Granite-Speech-5.0-2B 语音语言模型，基于 granite-4.0-1b-base 进行模态对齐训练，使用 39 万小时公开及合成音频数据。该模型支持英语、法语、德语、西班牙语、葡萄牙语和日语的多语言语音识别与双向语音翻译，相比前代在短语音识别准确率、噪声鲁棒性、长语音翻译和关键词偏置（最多 2000 词）方面有所提升。模型以 Apache 2.0 许可开源，面向企业语音处理场景。
正文节选
--- license: apache-2.0 language: - multilingual - en - fr - de - es - pt - ja base_model: - ibm-granite/granite-4.0-1b-base library_name: transformers pipeline_tag: automatic-speech-recognition --- # Granite-Speech-5.0-2B **Model Summary:** Granite Speech 5.0 2B is a compact and efficient speech-language model, specifically designed for multilingual automatic speech recognition (ASR) and bidirectional automatic speech translation (AST) for English, French, German, Spanish, Portuguese and Japan