انتقل إلى المحتوى

ibm-granitegranite-speech-4.1-2b

GraniteSpeech

Uses Granite Speech's instruction prompt with deterministic generation.

Automatic speech recognitionVoiceHub-nativegranite-speechParameters: 2.3BLanguages: en, fr +4Training: nativeLicense: Checkpoint-specific

Parameter metadata: Exact learned-parameter total for VoiceHub's audited native primary graph at the registered default selection; separately loaded auxiliary models are excluded.

Use this model Checkpoint
Resources
Upstream GitHub VoiceHub source Open in Colab

Usage

Complete the VoiceHub installation once, then run this repository-authored example. Model pages intentionally contain no package-install command.

This example is maintained against VoiceHub's public API; it is not copied from an upstream demo or package README.

Model-specific path: Uses Granite Speech's instruction prompt with deterministic generation.

Inputs and controls: Medical or regulated recordings still require domain review; model output is not a verified record.

from pathlib import Path

from voicehub import AutoModelForSpeechRecognition

AUDIO_FILE = Path("speech.wav")
if not AUDIO_FILE.is_file():
    raise FileNotFoundError(AUDIO_FILE)

model = AutoModelForSpeechRecognition.from_pretrained(
    'ibm-granite/granite-speech-4.1-2b',
    model_type='asr_granite_speech',
    device="cuda",
    lazy_load=True,
)
output = model.transcribe(
    AUDIO_FILE,
    prompt="Transcribe the recording faithfully in English.",
    do_sample=False,
)
print(output.text)
for segment in output.segments:
    print(segment.start, segment.end, segment.text, segment.confidence)

Use authorized recordings. Verify hardware needs and pin a revision in production.

Overview

asr_granite_speech is a VoiceHub automatic speech recognition integration. This page is generated from its registry contract. Open the asr_granite_speech Colab notebook.

Property Value
Task Automatic speech recognition
Architecture granite-speech
Runtime VoiceHub-native
Languages en, fr, de, es, pt, ja
Capabilities automatic-speech-recognition, multilingual, hotwords, translation, safetensors, fine-tuning, lora, voicehub-native, native-runtime
Reusable components —
Normalized output ASROutput

Language support

Supported language abbreviations

en, fr, de, es, pt, ja

These are ASR input languages; Italian and Mandarin are documented as translation targets only.

Paper and GitHub

Configuration

Load configuration without constructing the model:

from voicehub import AutoConfig

config = AutoConfig.for_model('asr_granite_speech')
print(config.model_type)
Property Value
Canonical model type asr_granite_speech
Configuration class GraniteSpeechASRConfig
Architecture class GraniteSpeechForSpeechRecognition

Processing

Create the registered processor without allocating model weights:

from voicehub import AutoProcessor

processor = AutoProcessor.from_pretrained(
    'ibm-granite/granite-speech-4.1-2b',
    model_type='asr_granite_speech',
)
print(type(processor).__name__)

Inference

The Usage example returns ASROutput through AutoModelForSpeechRecognition.

Input and output contract

Property Value
Readiness integrated-raw
Data architecture prompted-multimodal
Sample rate 16,000 Hz
Contract getter get_asr_dataset_spec('asr_granite_speech')
Variant Required fields One of Boundary Other rules
raw-audio audio text / transcription / transcript Source at most one: text / transcription / transcript; forbidden: language
granite-model-ready input_ids, attention_mask, input_features, input_features_mask, labels — Prepared —

Prompt-conditioned Granite Speech multimodal fine-tuning records. See the data workflow.

Training and optimization

Use available_optimization_passes() to discover reversible public passes. Unsupported runtime or hardware fails closed before mutation.

Training contract

Property Value
Support native
Family speech-sequence-to-sequence
Recipe single-phase
Default phase speech_recognition
Training checkpoint ibm-granite/granite-speech-4.1-2b
Native training graph yes
Phase Kind Components Required inputs Loss keys
speech_recognition objective model.encoder, model.projector, model.language_model input_ids, attention_mask, input_features, input_features_mask, labels loss

The integration accepts its declared source or prepared contract directly. Call model.validate_training_support() first, then follow the training workflow.

Checkpoints, provenance, license, and limitations

Property Value
Default checkpoint ibm-granite/granite-speech-4.1-2b
Hugging Face ID ibm-granite/granite-speech-4.1-2b
Repository availability verified through the Hugging Face model API on 2026-08-11; pin a revision before production use.
Checkpoint status Registry default; pin an immutable revision for production and reproducible evidence
Optional dependency extra Core package
Hardware and runtime Usage selects cuda; verify checkpoint-specific requirements
Real-checkpoint evidence Release evidence; a registry default alone is not execution evidence
Implementation voicehub.models.asr_granite_speech.modeling_asr_granite_speech.GraniteSpeechForSpeechRecognition
Configuration voicehub.models.asr_granite_speech.configuration_asr_granite_speech.GraniteSpeechASRConfig
Source provenance voicehub/architectures/granite_speech/SOURCE.json
License Checkpoint-specific

No VoiceHub-specific license override is registered. Verify the checkpoint and upstream source terms before use.

Confirm the checkpoint revision, access terms, provenance, and license.

Limitations

  • No integration-specific checkpoint limitation is registered. Verify the selected checkpoint revision and its documented runtime requirements.
  • Validate memory, precision, and optional dependencies on the target system.
  • Public optimizations fail closed when the runtime or hardware cannot satisfy their validation contract; an unavailable pass is not reported as applied.
  • Contract tests do not replace the linked released-checkpoint evidence.

Public API

Use the stable configuration, processor, and task-model facades below.

Configuration

GraniteSpeechASRConfig

View source

GraniteSpeechASRConfig(**config_kwargs)

Parameters

  • **config_kwargs — Configuration fields validated by GraniteSpeechASRConfig.

Model

GraniteSpeechForSpeechRecognition

View source

AutoModelForSpeechRecognition.from_pretrained(
    pretrained_model_name_or_path,
    *,
    model_type='asr_granite_speech',
    config=None,
    **model_kwargs,
)

Parameters

  • pretrained_model_name_or_path — Hub ID or compatible local directory.
  • model_type — Canonical model type; use 'asr_granite_speech'.
  • config — Optional preloaded GraniteSpeechASRConfig instance.
  • **model_kwargs — Model-specific loading arguments.
from voicehub import get_model_spec

spec = get_model_spec('asr_granite_speech')
print(spec.display_name, spec.task.value)
Purpose Public object
Discover get_model_spec('asr_granite_speech')
Load and run AutoModelForSpeechRecognition
Configure GraniteSpeechASRConfig
Process AutoProcessor
Model implementation GraniteSpeechForSpeechRecognition
Normalized output ASROutput
Training contract get_training_spec('asr_granite_speech')
Optimization lifecycle available_optimization_passes, apply_optimization_plan, optimization_manifest, restore_optimization_plan

See all model guides, inference, and the training matrix.