Open4bits/whisper-tiny-f16

1---2language:3- en4- zh5- de6- es7- ru8- ko9- fr10- ja11- pt12- tr13- pl14- ca15- nl16- ar17- sv18- it19- id20- hi21- fi22- vi23- he24- uk25- el26- ms27- cs28- ro29- da30- hu31- ta32- 'no'33- th34- ur35- hr36- bg37- lt38- la39- mi40- ml41- cy42- sk43- te44- fa45- lv46- bn47- sr48- az49- sl50- kn51- et52- mk53- br54- eu55- is56- hy57- ne58- mn59- bs60- kk61- sq62- sw63- gl64- mr65- pa66- si67- km68- sn69- yo70- so71- af72- oc73- ka74- be75- tg76- sd77- gu78- am79- yi80- lo81- uz82- fo83- ht84- ps85- tk86- nn87- mt88- sa89- lb90- my91- bo92- tl93- mg94- as95- tt96- haw97- ln98- ha99- ba100- jw101- su102tags:103- audio104- automatic-speech-recognition105- hf-asr-leaderboard106- open4bits107widget:108- example_title: Librispeech sample 1109  src: https://cdn-media.huggingface.co/speech_samples/sample1.flac110- example_title: Librispeech sample 2111  src: https://cdn-media.huggingface.co/speech_samples/sample2.flac112model-index:113- name: whisper-tiny114  results:115  - task:116      name: Automatic Speech Recognition117      type: automatic-speech-recognition118    dataset:119      name: LibriSpeech (clean)120      type: librispeech_asr121      config: clean122      split: test123      args:124        language: en125    metrics:126    - name: Test WER127      type: wer128      value: 7.54129  - task:130      name: Automatic Speech Recognition131      type: automatic-speech-recognition132    dataset:133      name: LibriSpeech (other)134      type: librispeech_asr135      config: other136      split: test137      args:138        language: en139    metrics:140    - name: Test WER141      type: wer142      value: 17.15143  - task:144      name: Automatic Speech Recognition145      type: automatic-speech-recognition146    dataset:147      name: Common Voice 11.0148      type: mozilla-foundation/common_voice_11_0149      config: hi150      split: test151      args:152        language: hi153    metrics:154    - name: Test WER155      type: wer156      value: 141157pipeline_tag: automatic-speech-recognition158license: apache-2.0159base_model:160- openai/whisper-tiny161---162 163# Open4bits / Whisper Tiny FP16164 165This repository provides the **Whisper Tiny model converted to FP16 (float16) precision**, published by Open4bits to enable highly efficient inference with minimal memory usage.166 167The underlying Whisper model and architecture are **owned by OpenAI**. This repository contains only a precision-converted version of the original model weights.168 169The model is designed for fast, lightweight multilingual speech-to-text tasks and is well suited for resource-constrained environments.170 171---172 173## Model Overview174 175Whisper is a sequence-to-sequence transformer model developed by OpenAI for automatic speech recognition and speech translation.  176This release uses the **Tiny** variant, prioritizing speed and low memory usage while preserving the original architecture.177 178---179 180## Model Details181 182- **Architecture:** Whisper Tiny  183- **Parameters:** ~37.85 million  184- **Precision:** float16 (FP16)  185- **Task:** Automatic Speech Recognition (ASR)  186- **Languages:** Multilingual  187- **Weight tying:** Preserved  188- **Compatibility:** Hugging Face Transformers, PyTorch  189 190Compared to larger Whisper variants, this model offers significantly faster inference and lower VRAM requirements, with reduced accuracy in some scenarios.191 192---193 194## Intended Use195 196This model is intended for:197- Fast speech-to-text transcription198- Lightweight and real-time ASR applications199- Edge or low-resource deployments200- Research and prototyping201 202---203 204## Limitations205 206* Lower transcription accuracy compared to larger Whisper variants207* Performance depends on audio quality, language, and accent208* Not fine-tuned for domain-specific or noisy audio209 210---211 212## License213 214This model is released under the **Apache License 2.0**.215The original Whisper model and associated intellectual property are owned by OpenAI.216 217---218 219## Support220 221If you find this model useful, please consider supporting the project.222Your support helps us continue releasing and maintaining high-quality open models.223Support us with a heart.