Automatic Speech Recognition
Transformers
Safetensors
Danish
cohere_asr
audio
speech-recognition
transcription
danish
hf-asr-leaderboard
custom_code
Instructions to use syvai/hviske-v5.3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use syvai/hviske-v5.3 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="syvai/hviske-v5.3", trust_remote_code=True)# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq processor = AutoProcessor.from_pretrained("syvai/hviske-v5.3", trust_remote_code=True) model = AutoModelForSpeechSeq2Seq.from_pretrained("syvai/hviske-v5.3", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
How are people using this to generate subtitles?
#7
by GeoMaciolek - opened
I'm curious how people are using this in the real world to generate subtitles. I've had medium success with the script including in this script in an earlier comment thread.
I'd been hoping to use a tool like e.g. SubtitleEdit (which at https://www.nikse.dk/subtitleedit is apparently Danish!) but the best built-in model support seems to be Whisper V3 Large, despite supporting CrispASR (but not in a flexible way.)
I've tried to use subtitleedit with hviske-v5.3 (on vLLM via an OpenAI-Compatible endpoint), without success. I'm curious- I can't be the only person wanting to use a highly-rated Danish ASR engine to generate subtitles, so I figured I should ask!