3AM Marketer
AI

IBM Granite Speech 5.0 Delivers 5.06% WER in New Update

IBM's Granite Speech collection delivers an Apache 2.0 alternative to Whisper, pairing a 5.06% English WER with documented VRAM and runtime profiles across speech models.

average English WER5.06% WER
5 sources, checked 22 Oct 2026.

IBM released three updates to its Granite Speech collection, introducing the granite-speech-5.0-2b model alongside revised checkpoints for granite-speech-5.0-470m-turboctc and granite-speech-4.1-2b-plus. The granite-speech-5.0-2b release ranks second among top open-weight speech recognition systems with a 5.06% average English word error rate on the Open ASR Leaderboard, trained across 390,000 hours of public audio supplemented by synthetic data.1

For engineering teams deploying local AI models, automatic speech recognition often involves balancing word accuracy against hardware overhead. The release expands beyond English by transcribing French, German, Spanish, Portuguese, and Japanese, while translating to and from English in those languages, as well as handling English translation to Italian and Mandarin.1

Transcription throughput and 19,870 RTFx speeds

Execution latency varies substantially between speech architectures. The granite-speech-5.0-470m-turboctc checkpoint reached 19,870 RTFx on the Open ASR Leaderboard, up from roughly 12,600 RTFx at launch, after compiling a single Conformer block and reusing its graph across the stack. The Turbo CTC model provides multilingual transcription across seven languages: English, Japanese, Spanish, German, French, Italian, and Portuguese.1

By comparison, SenseVoiceSmall uses a non-autoregressive end-to-end architecture supporting Mandarin, Cantonese, English, Japanese, and Korean. At a similar parameter scale, SenseVoiceSmall runs more than 5 times faster than Whisper-Small and 15 times faster than Whisper-Large.2

Keyword biasing lists and 7% diarization gains

Granite Speech 5.0 2B adds keyword biasing support for lists of up to 2,000 terms in every supported language to steer recognition toward product names, people, and acronyms. It also outputs punctuation and capitalization on request, recording capitalization F1 scores between 89.7 and 99.5 across evaluation benchmarks.1

The updated granite-speech-4.1-2b-plus checkpoint focuses on long-form audio and speaker attribution, dropping both its word diarization error rate and speaker-attributed WER by 7%. Its word diarization error rate fell from 4.40 to 4.10, while speaker-attributed transcripts decreased from 30.50 to 28.48 cpWER. Long-form WER dropped from 13.53 to 10.22, average timestamp shift declined from 210.86 to 98.54, and biased-word WER on 20-term lists decreased from 11.79 to 6.48.1

  • Word diarization (before)4.4
  • Word diarization (after)4.1
  • Speaker cpWER (before)30.5
  • Speaker cpWER (after)28.48
  • Long-form WER (before)13.53
  • Long-form WER (after)10.22
  • Biased-word WER (before)11.79
  • Biased-word WER (after)6.48
Granite Speech 4.1 2B Plus error rate and latency reductions across task benchmarks.1

Hardware footprints across speech synthesis models

Memory footprints dictate whether a speech pipeline can share an accelerator with self-hosted AI agents or requires dedicated compute.345

Hardware memory requirements and licensing across open-weight speech models.345
ModelTaskHardware MinimumLicence
KokoroTTSCPU onlyApache 2.0
More on Kokoro
Licence
Apache 2.0
Whisper tinyASR~1 GB VRAMMIT
More on Whisper tiny
Licence
MIT
Whisper smallASR~2 GB VRAMMIT
More on Whisper small
Licence
MIT
Whisper mediumASR~5 GB VRAMMIT
More on Whisper medium
Licence
MIT
Whisper turboASR~6 GB VRAMMIT
More on Whisper turbo
Licence
MIT
Whisper largeASR~10 GB VRAMMIT
More on Whisper large
Licence
MIT
Qwen3-TTSTTS6 GB+ VRAMApache 2.0
More on Qwen3-TTS
Licence
Apache 2.0
VoxtralTTS16 GB+ VRAMCC BY-NC 4.0
More on Voxtral
Licence
CC BY-NC 4.0
Comparison table of open-source text-to-speech models, parameters, licenses, and hardware requirements.
Screenshot of texttolab.com, captured 2026-10-22

Runtime compatibility and licence terms

Deployment viability often depends on software runtime integration and legal terms. Granite-speech-5.0-2b ships under an Apache 2.0 licence with native support in Transformers, vLLM, llama.cpp, and mlx-audio for Apple Silicon. OpenAI distributes Whisper model weights and codebase under the MIT License.13

In speech synthesis, licensing divides production commercial deployments. Kokoro and Qwen3-TTS run under Apache 2.0, and Bark TTS operates under the MIT licence for commercial use. In contrast, Voxtral enforces a restrictive CC BY-NC 4.0 licence, excluding commercial workflows regardless of available VRAM.45

Teams choosing speech pipelines should audit licensing terms before selecting architectures. When permissive commercial use and structured keyword biasing are required on standard local accelerators, Granite Speech 5.0 and Whisper provide permissive baselines, while non-commercial weights like Voxtral remain restricted to research environments.134