Granite Speech 5.0: Transcribe, Translate, and Attribute more Accurately and Faster

本文为原文前 6,000 字符的节选翻译，完整内容请查看原文。

Back to Articles Granite Speech 5.0: Transcribe, Translate, and Attribute more Accurately and Faster Enterprise Article Published October 21, 2026.

返回文章：Granite Speech 5.0：更准确、更快速地进行转录、翻译和归因。企业文章，发布于 2026 年 10 月 21 日。

We’re excited to share three updates to IBM’s Granite Speech collection: a new speech-language model, granite-speech-5.0-2b, and revised versions of two models the community already uses, granite-speech-5.0-470m-turboctc and granite-speech-4.1-2b-plus. Together they cover the full range of enterprise speech work: transcription and translation with high accuracy, extremely fast transcription at the edge, and recordings where who spoke, and when, matter as much as what was said.

我们很高兴分享 IBM Granite Speech 系列的三项更新：一个新的语音语言模型 granite-speech-5.0-2b，以及社区已在使用的两个模型的修订版本：granite-speech-5.0-470m-turboctc 和 granite-speech-4.1-2b-plus。它们共同涵盖了企业语音工作的全方位需求：高精度的转录与翻译、极速的边缘端转录，以及对“谁在何时发言”与“说了什么”同样重要的录音处理。

Granite Speech 5.0 2B: more accurate, more robust, more control. granite-speech-5.0-2b is the successor to granite-speech-4.1-2b at the same parameter count. It transcribes English, French, German, Spanish, Portuguese, and Japanese, and translates to and from English in those languages, plus English to Italian and Mandarin. It was trained on 390,000 hours of public audio, plus synthetic data built for Japanese ASR, keyword biasing, and translation.

Granite Speech 5.0 2B：更准确、更稳健、更易控。granite-speech-5.0-2b 是 granite-speech-4.1-2b 的继任者，参数量保持不变。它支持英语、法语、德语、西班牙语、葡萄牙语和日语的转录，并支持这些语言与英语之间的互译，此外还支持英语到意大利语和普通话的翻译。该模型在 39 万小时的公共音频数据上进行了训练，并结合了为日语自动语音识别（ASR）、关键词偏置和翻译构建的合成数据。

What’s new compared with 4.1 2B Speech: More accurate, more robust English ASR: Higher accuracy on English short-form benchmarks, and better handling of noise, reverberation, and distant-microphone speech. Language ID prompting: Name the input language in the prompt to steer multilingual transcription. Better long-form and conversational translation: Translation holds up for longer with more natural speech, not just short read sentences.

与 4.1 2B 相比的新特性：更准确、更稳健的英语 ASR：在英语短格式基准测试中具有更高的准确性，并能更好地处理噪声、混响和远场麦克风语音。语言 ID 提示：在提示词中指定输入语言以引导多语言转录。更好的长文本和对话翻译：翻译在更长的语境下表现更稳定，语音更自然，不仅限于简短的朗读句子。

Granite Speech 5.0 Turbo CTC: now multilingual, more robust, and even faster. When we introduced Turbo CTC in August, it was an English-only, 470M-parameter encoder built for speed. This revision keeps that footprint and adds the three things users asked for most: Multilingual transcription: English, Japanese, Spanish, German, French, Italian, and Portuguese. Native punctuation and capitalization: Transcripts come out ready to read, with no separate punctuation or truecasing step.

Granite Speech 5.0 Turbo CTC：现已支持多语言，更稳健且速度更快。我们在 8 月推出 Turbo CTC 时，它是一个仅支持英语、拥有 4.7 亿参数且专为速度而构建的编码器。此次修订保留了原有的规模，并增加了用户最需要的三个功能：多语言转录（英语、日语、西班牙语、德语、法语、意大利语和葡萄牙语）；原生标点和大小写（转录结果直接可读，无需额外的标点或大小写修正步骤）。

Granite Speech 4.1 2B Plus: better at who said what, and when. The updated granite-speech-4.1-2b-plus is tuned for the recordings enterprises actually have: meetings, support calls, and interviews with several speakers running long. It keeps the 4.1 name because the architecture, prompt format, and API are unchanged, so it’s a drop-in replacement for the current model with improved weights.

Granite Speech 4.1 2B Plus：更擅长识别“谁在何时说了什么”。更新后的 granite-speech-4.1-2b-plus 针对企业实际录音进行了调优，包括会议、支持通话以及多发言人的长时访谈。它保留了 4.1 的名称，因为其架构、提示格式和 API 均未改变，因此它是当前模型在权重优化后的直接替代品。