Search FoxAI
AI Models 535
- faster-whisper-base-int8↗
rhasspy
- whisper-tiny-fp16-ov↗
OpenVINO
- kotoba-whisper-bilingual-v1.0↗
kotoba-tech
_Kotoba-Whisper-Bilingual_ is a collection of distilled Whisper models trained for - **Japanese ASR** - **English ASR** - **Speech-to-text translation (Japanese -> English)** - **Speech-to-text translation (English -> Japanese)** developed through the collaboration bewteen Asahi Ushio and Kotoba Technologies. Following the original work of distil-whisper (Robust Knowledge Distillation via Large-Scale Pseudo Labelling), we employ OpenAI's Whisper large-v3 as the teacher model for Japanese and English ASR, while we translate the transcription into English and Japanese by external LLM to obtain training dataset for speech-to-text translation. We employ ReazonSpeech for Japanese ASR and Japanese speech to English text translation, and Multilingual LibriSpeech for English ASR and English speech to Japanese text translation. Kotoba-whisper-bilingual's loss objective consists of cross-entropy on both of ASR and translation tasks, while KL divergence loss only for ASR task. The student model consists the full encoder of the teacher large-v3 model and the decoder with two layers initialized from the first and last layer of the large-v3 model. As kotoba-whisper uses the same architecture as distil-whisper/distil-large-v3, it inherits the benefit of the improved latency compared to openai/whisper-large-v3 (**6.3x faster than large-v3**, see the table below taken from distil-whisper/distil-large-v3).
- afrispeech-whisper-medium-all↗
intronhealth
- kyrgyz-whisper-medium↗
nineninesix
- faster-whisper-large-v3-edacc-commonvoice-l2arctic-v3↗
2snem6
- faster-whisper-large-v3-turbo-int8-ct2↗
Zoont
This model is converted from openai/whisper-large-v3-turbo to the CTranslate2 format using INT8 quantization, primarily for use with faster-whisper
- whisper-ja-760M↗
efwkjn
- whisper-turbo-ksc2↗
abilmansplus
- whisper-large-v3-turbo-danish↗
thorhojhus