Update README.md
Browse files
README.md
CHANGED
@@ -9,8 +9,8 @@ base_model: biodatlab/whisper-th-large-combined
|
|
9 |
|
10 |
whisper-th-large-ct2 is the CTranslate2 format of [biodatlab/whisper-th-large-combined](https://huggingface.co/biodatlab/whisper-th-large-combined), comparable with [WhisperX](https://github.com/m-bain/whisperX) and [faster-whisper](https://github.com/SYSTRAN/faster-whisper), which enables:
|
11 |
|
12 |
-
- ⚡️ Batched inference for 70x real-time transcription using Whisper large-v2.
|
13 |
-
- 🪶 A faster-whisper backend, requiring
|
14 |
- 🎯 Accurate word-level timestamps using wav2vec2 alignment.
|
15 |
- 👯♂️ Multispeaker ASR using speaker diarization from pyannote-audio (includes speaker ID labels).
|
16 |
- 🗣️ VAD preprocessing, reducing hallucinations and allowing batching with no WER degradation.
|
|
|
9 |
|
10 |
whisper-th-large-ct2 is the CTranslate2 format of [biodatlab/whisper-th-large-combined](https://huggingface.co/biodatlab/whisper-th-large-combined), comparable with [WhisperX](https://github.com/m-bain/whisperX) and [faster-whisper](https://github.com/SYSTRAN/faster-whisper), which enables:
|
11 |
|
12 |
+
- ⚡️ Batched inference for **70x** real-time transcription using Whisper large-v2.
|
13 |
+
- 🪶 A faster-whisper backend, requiring **<8GB GPU memory** for large-v2 with beam_size=5.
|
14 |
- 🎯 Accurate word-level timestamps using wav2vec2 alignment.
|
15 |
- 👯♂️ Multispeaker ASR using speaker diarization from pyannote-audio (includes speaker ID labels).
|
16 |
- 🗣️ VAD preprocessing, reducing hallucinations and allowing batching with no WER degradation.
|