--- title: Granite Speech 5.0: Transcribe, Translate, and Attribute... description: IBM released three updated Granite Speech models for enterprise transcription and translation. The new granite-speech-5.0-2b transcribes six languages and... canonical: https://daily.dev/posts/granite-speech-5-0-transcribe-translate-and-attribute-more-accurately-and-faster-nyvlfrpof twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Granite Speech 5.0: Transcribe, Translate, and Attribute more Accurately and Faster | daily.dev og:description: IBM released three updated Granite Speech models for enterprise transcription and translation. The new granite-speech-5.0-2b transcribes six languages and... og:url: https://daily.dev/posts/granite-speech-5-0-transcribe-translate-and-attribute-more-accurately-and-faster-nyvlfrpof og:image: https://api.daily.dev/og/posts/nyVLFrpoF.png og:image:alt: Granite Speech 5.0: Transcribe, Translate, and Attribute more Accurately and Faster og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Granite Speech 5.0: Transcribe, Translate, and Attribute more Accurately and Faster **[Hugging Face](https://daily.dev/sources/huggingface)** · 5 min read · 0 upvotes · 0 comments ## Summary IBM released three updated Granite Speech models for enterprise transcription and translation. The new granite-speech-5.0-2b transcribes six languages and translates several of them, trained on 390,000 hours of audio, and ranks #2 among open-weight ASR models with 5.06% average English WER. It supports keyword lists up to 2,000 terms. The updated 470M Turbo CTC model is now multilingual and hit 19,870 RTFx, about 1.6x faster than its August launch. The updated granite-speech-4.1-2b-plus improves meeting transcription: WDER and cpWER each drop 7%, timestamp shift is cut by more than half, and long-form WER drops 24%. All models ship under Apache 2.0 with Transformers support. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How fast is IBM's granite-speech-5.0-470m-turboctc model compared to its earlier version? It reached 19,870 RTFx on the Open ASR Leaderboard, up from roughly 12,600 RTFx at launch, about 1.6x faster at the same accuracy level. The speedup came from compiling a single Conformer block and reusing its graph across the stack, cutting warmup time. It is a 470M-parameter English-only-turned-multilingual encoder built for speed. _Developers tracking ASR speed and accuracy tradeoffs can follow model updates like this on daily.dev._ ### Which languages does granite-speech-5.0-2b support for transcription and translation? It transcribes English, French, German, Spanish, Portuguese, and Japanese. It translates to and from English in those languages, plus English to Italian and Mandarin. The model was trained on 390,000 hours of public audio plus synthetic data built for Japanese ASR, keyword biasing, and translation, and supports keyword lists up to 2,000 terms. _Teams choosing a multilingual ASR model can keep up with releases like this on daily.dev._ ### How much did speaker attribution accuracy improve in granite-speech-4.1-2b-plus? Word diarization error rate (WDER) dropped 7%, from 4.40 to 4.10, and speaker-attributed WER (cpWER) dropped 7% overall, from 30.50 to 28.48. Average timestamp shift across all speech was cut by more than half, from 210.86 to 98.54, and long-form WER fell 24%, from 13.53 to 10.22. _Developers building multi-speaker transcription tools can track diarization benchmarks like these on daily.dev._ ## Similar posts on daily.dev - [Extremely Fast and Accurate Transcription with Granite Speech 5.0 Turbo CTC](https://daily.dev/posts/extremely-fast-and-accurate-transcription-with-granite-speech-5-0-turbo-ctc-ivnplcd09) · Hugging Face · 1 upvotes · 0 comments - [Granite 4.0 1B Speech: Compact, Multilingual, and Built for the Edge](https://daily.dev/posts/granite-4-0-1b-speech-compact-multilingual-and-built-for-the-edge-djlhogrk3) · Hugging Face · 22 upvotes · 0 comments - [Granite Switch 4.1: Chat, Decisions, and Speech from One Endpoint](https://daily.dev/posts/granite-switch-4-1-chat-decisions-and-speech-from-one-endpoint-0ig1nmhow) · Hugging Face · 1 upvotes · 1 comments --- Tags: [#speech-recognition](https://daily.dev/tags/speech-recognition) [View this post on daily.dev](https://daily.dev/posts/granite-speech-5-0-transcribe-translate-and-attribute-more-accurately-and-faster-nyvlfrpof) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Granite Speech 5.0: Transcribe, Translate, and Attribute more Accurately and Faster","url":"https://daily.dev/posts/granite-speech-5-0-transcribe-translate-and-attribute-more-accurately-and-faster-nyvlfrpof","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/granite-speech-5-0-transcribe-translate-and-attribute-more-accurately-and-faster-nyvlfrpof"},"datePublished":"2026-10-21T14:58:32.692Z","dateModified":"2026-10-21T15:20:29.679Z","description":"IBM released three updated Granite Speech models for enterprise transcription and translation. The new granite-speech-5.0-2b transcribes six languages and...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/671b43382c78001f0080422993405628?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/671b43382c78001f0080422993405628?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Hugging Face","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Hugging Face","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/f1f55c67d81a4330acf5b90b26b0c8e1","url":"https://daily.dev/sources/huggingface"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/granite-speech-5-0-transcribe-translate-and-attribute-more-accurately-and-faster-nyvlfrpof","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"speech-recognition","timeRequired":"PT5M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Hugging Face","item":"https://daily.dev/sources/huggingface"},{"@type":"ListItem","position":3,"name":"Granite Speech 5.0: Transcribe, Translate, and Attribute more Accurately and Faster"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/granite-speech-5-0-transcribe-translate-and-attribute-more-accurately-and-faster-nyvlfrpof#faq","mainEntity":[{"@type":"Question","name":"How fast is IBM's granite-speech-5.0-470m-turboctc model compared to its earlier version?","acceptedAnswer":{"@type":"Answer","text":"It reached 19,870 RTFx on the Open ASR Leaderboard, up from roughly 12,600 RTFx at launch, about 1.6x faster at the same accuracy level. The speedup came from compiling a single Conformer block and reusing its graph across the stack, cutting warmup time. It is a 470M-parameter English-only-turned-multilingual encoder built for speed. Developers tracking ASR speed and accuracy tradeoffs can follow model updates like this on daily.dev."}},{"@type":"Question","name":"Which languages does granite-speech-5.0-2b support for transcription and translation?","acceptedAnswer":{"@type":"Answer","text":"It transcribes English, French, German, Spanish, Portuguese, and Japanese. It translates to and from English in those languages, plus English to Italian and Mandarin. The model was trained on 390,000 hours of public audio plus synthetic data built for Japanese ASR, keyword biasing, and translation, and supports keyword lists up to 2,000 terms. Teams choosing a multilingual ASR model can keep up with releases like this on daily.dev."}},{"@type":"Question","name":"How much did speaker attribution accuracy improve in granite-speech-4.1-2b-plus?","acceptedAnswer":{"@type":"Answer","text":"Word diarization error rate (WDER) dropped 7%, from 4.40 to 4.10, and speaker-attributed WER (cpWER) dropped 7% overall, from 30.50 to 28.48. Average timestamp shift across all speech was cut by more than half, from 210.86 to 98.54, and long-form WER fell 24%, from 13.53 to 10.22. Developers building multi-speaker transcription tools can track diarization benchmarks like these on daily.dev."}}]} ```