Skip to main content
Audio modelOpen weights

Granite Speech 5.0 470M TurboCTC

IBM

Released
August 25, 2026
Data date
October 3, 2026

Granite Speech 5.0 470M TurboCTC is a 470-million-parameter model for English speech recognition. Its Conformer architecture uses non-autoregressive CTC decoding and targets edge devices, laptops, and mobile hardware.

The checkpoint is available locally under Apache-2.0. Its role is audio to text, not audio understanding with a spoken response.

Specifications and access

SpecificationValue and source
Model class
Speech recognitionSource
Access
Local weightsSource
Input
English audioSource
Output
TranscriptSource
Checkpoint
ibm-granite/granite-speech-5.0-470m-turboctcSource
License
Apache-2.0Source
Architecture
Conformer with non-autoregressive CTC decodingSource
Deployment
Designed for edge devices, laptops, and mobile devicesSource