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

IBM has released three updates to its Granite Speech collection, including a new speech-language model, granite-speech-5.0-2b, and updated versions of granite-speech-5.0-470m-turboctc and granite-speech-4.1-2b-plus. These models aim to enhance enterprise speech tasks such as transcription, translation, and speaker attribution. Improvements are noted in areas like average timestamp shift (AAS) for all speech (98.54), keyword biasing (6.48 Biased-word WER), and speaker attribution (4.10 WDER).