Daniel Schnell

Is there a simple way to say: transition from each state for specific event ?

2

How can I use a simple way to say: transition always to a specific state with a specific event, without having to specify this event for each state of the...

[Bug] Notebooks don't work on branch dev

5

### Describe the bug E.g. `notebooks/ExtractTTSpectrogram.ipynb`:  `notebooks/Tutorial_2_train_your_first_TTS_model.ipynb`:  ### To Reproduce Run the notebooks ### Expected behavior All notebooks run against current default branch (`dev`). ### Logs _No response_...

Link error

1

If clicking on the demo app on Heroku: 

If no file is given, download all assets

1

The error is somewhat misleading, if no `file:` key is given: ``` /home/runner/work/_actions/dsaltares/fetch-gh-release-asset/master/dist/index.js:2050 throw new Error(`Input required and not supplied: ${name}`); ``` I rather supposed that all files would be...

I have a schema with an array of oneOf $refs : ``` ruby schema='{ "$schema": "http://json-schema.org/draft-04/schema#", "title": "demo", "type": "object", "properties": { "formats": { "type": "array", "items": { "$ref": "#/definitions/format"...

Bad timestamp prediction with some finetuned Whisper models

9

I see the following incorrect transcriptions, when running my tests with the fine-tuned model [language-and-voice-lab/whisper-large-icelandic-62640-steps-967h](https://huggingface.co/language-and-voice-lab/whisper-large-icelandic-62640-steps-967h) : ```json "segments": [ { "id": 0, "seek": 0, "start": 0.0, "end": 30.0, "text": "afbrot...

Single word / part of sentence correction

2

I want to use Greynir-Correct for correction of non-whole sentences, i.e. in extreme cases single words. What method or options should I use to make that possible ? Currently, when...

Strange behavior of tokenize(.., only_ci=True)

1

The following snippet gives inconsistent results: ```python from reynir_correct import tokenize texts = ["Skúta", "300 ára gömul írsk skúta fundin við Suður-Noreg" ] for t in texts: g = tokenize(t,...

If there are multiple possibilities to correct a word, e.g. from POV of the same Levenshtein distance, is it possible to get back all possible corrections greynir-correct knows ? If...

Does model output phoneme-level timing info ?

1

Hi, thanks for your work. I'd be interested, if the model also provides phoneme-level timing information at inference ?