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

AI小蝌蚪AI 前沿📡 Hugging Face Blog2026-10-22557 阅读💛 167 收藏
Back to Articles

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

Enterprise Article
Published October 21, 2026
George Saon's avatar
George Saon
gsaon
IBM Granite's avatar ibm-granite
Brian Kingsbury's avatar
Brian Kingsbury
bedkingsbury
IBM Granite's avatar ibm-granite
Ron Hoory's avatar
Ron Hoory
rhoory
IBM Granite's avatar ibm-granite
Vishal Sunder's avatar
Vishal Sunder
vishalsunder
IBM Granite's avatar ibm-granite
Zvi Kons's avatar
Zvi Kons
konszvi
IBM Granite's avatar ibm-granite
Jeff Kuo's avatar
Jeff Kuo
ibm-jeffkuo
IBM Granite's avatar ibm-granite
Samuel Thomas's avatar
Samuel Thomas
sthomas12
IBM Granite's avatar ibm-granite
Avihu Dekel's avatar
Avihu Dekel
Avihu
IBM Granite's avatar ibm-granite
Hagai Aronowitz's avatar
Hagai Aronowitz
hagaia
IBM Granite's avatar ibm-granite
Takashi Fukuda's avatar
Takashi Fukuda
fukuda1
IBM Granite's avatar ibm-granite
Tohru Nagano's avatar
Tohru Nagano
tohru3
IBM Granite's avatar ibm-granite
Sashi Novitasari's avatar
Sashi Novitasari
sasnov
IBM Granite's avatar ibm-granite
Masayuki Suzuki's avatar
Masayuki Suzuki
szuk
IBM Granite's avatar ibm-granite
Madison Lee's avatar
Madison Lee
kristunlee
IBM Granite's avatar ibm-granite

We're excited to share three updates to IBM's Granite Speech collection: a new speech-language model, , and revised versions of two models the community already uses, and . 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.

Model Status What's new Best for
granite-speech-5.0-2b New More accurate, noise- and reverb-robust automatic speech recognition (ASR), language-ID prompting, 2,000-term keyword biasing, better long-form translation High-accuracy multilingual transcription and translation
granite-speech-5.0-470m-turboctc Updated Multilingual, native punctuation and capitalization, and more robust to noise and reverb High-throughput and on-device transcription
granite-speech-4.1-2b-plus Updated Improved speaker attribution, timestamps, keyword biasing, and long-speech performance Meetings, calls, and other multi-speaker recordings

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.

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.
  • Keyword lists up to 2,000 terms: Bias transcription and translation toward product names, people, and acronyms, in every supported language.
  • Punctuation and capitalization on request: Ask for formatted output in the prompt; capitalization F1 ranges from 89.7 to 99.5 across our test sets.

The model ships under Apache 2.0 with support in Transformers, vLLM, llama.cpp, and mlx-audio for Apple Silicon.

granite-speech-5.0-2b average English WER compared with top open-weight ASR models

Chart 1: granite-speech-5.0-2b ranks #2 among top open-weight ASR models on average English WER (5.06%). Lower is better. Results are official, as of 21 October 2026, and averaged over the Open ASR Leaderboard’s public English short-form test sets.

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.
  • More robust in difficult acoustics: Training that explicitly targets reverberant audio makes Turbo CTC more robust in echoey rooms and far-field recordings.

It's also faster. By compiling a single Conformer block and reusing its graph across the stack, we cut warmup time and pushed Turbo CTC to 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.

Chart 2: Average WER vs. RTFx on the Open ASR Leaderboard (English short-form, 9 public test sets, as of October 2026). Lower WER and higher RTFx are better; RTFx is on a log scale.

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. This revision improves four areas:

  • Speaker attribution: More accurate labeling of who is speaking. Word diarization error rate (WDER) drops 7%, and speaker-attributed WER (cpWER) drops 7% overall.
  • Timestamps: Average timestamp shift (AAS) is cut by more than half, yielding tighter timing in captions, search, and review.
  • Long-speech performance: WER on long-form test sets drops 24%.
  • Keyword biasing: Errors on biased keywords fall 45% with 20-term keyword lists.
Capability Metric Previous release Updated model
Timestamps, all speech AAS (average timestamp shift) 210.86 98.54
Timestamps, short speech AAS (average timestamp shift) 29.27 26.20
Keyword biasing Biased-word WER, 20-term lists 11.79 6.48
Long-form speech WER, long-form test sets 13.53 10.22
Speaker attribution WDER 4.40 4.10
Speaker-attributed transcripts cpWER 30.50 28.48

Table 1: All metrics are error rates, so lower is better. For more detailed results, see the model card.

Get started

Not sure which model to pick? Start here:

  • Need high transcription accuracy, translation, or keyword biasing? Use .
  • Need raw speed or on-device transcription? Use , and try it in the browser with our WebGPU streaming demo.
  • Working with meetings or multi-speaker calls? Use .

These models are all released under an Apache 2.0 license with native support in Transformers. Full evaluation results, architecture details, training data, and usage examples are on each model card. We recommend pairing them with Granite Guardian for production deployments that need additional risk detection.

Give them a try and let us know what you think!

文章评论(12)

白向阳25 天前

不错不错,已加入书签。

回复
月沐雨25 天前

这个观点很中肯,深有同感。

回复
竹听风25 天前

收藏了,以后慢慢研究。

回复
蓝光25 天前

实话,说的不明不白

回复
蓝临风25 天前

写得挺用心的,支持一下。

回复
山问津25 天前

这个观点很中肯,深有同感。

回复
风倚栏25 天前

内容翔实,正好需要,先收藏再看。

回复
竹听风25 天前

楼主辛苦了,内容很有参考价值。

回复
龙文博25 天前

这篇文章分析得很透彻,收藏了!

回复
星知秋25 天前

刚好最近在找这方面的资料,太及时了。

回复
空向阳25 天前

写得挺用心的,支持一下。

回复
晨沐雨24 天前

收藏了,以后慢慢研究。

回复