-
Thanks for publishing the code and basic training instructions!
## Environment
**Datasets:** (9,063 speakers)
- LibriTTS (train-other-500)
- VoxCeleb1
- VoxCeleb2
- OpenSLR (42-44, 61-66, 69…
-
Hello!
I have wav2letter running and able to transcribe wav files which is fantastic. The results aren't very accurate though, but I believe this is as a result of using Australian accents while th…
-
Hi,
I'm testing the SOTA/2019 recipes on my own dataset and it seems that the TDS CTC recipe works for the data. However, when I use the TDS S2S recipe the WER is over 100 and is the same for the f…
-
The three example fil files throw an error when used with heimdall-astro (I tried using heimdall for verification)
-
Is it possible to compute log Mel-filterbank energy features like it is provided by the function logfbank in python_speech_features?
https://python-speech-features.readthedocs.io/en/latest/#python_sp…
-
**Submitting author:** @telegraphic (Danny Price)
**Repository:** https://github.com/UCBerkeleySETI/blimpy
**Version:** 1.4.1
**Editor:** @xuanxu
**Reviewer:** @garrettj403, @zhampel
**Archive:** …
-
Hi, I am in need of help. I have downloaded a pre-trained model which I am trying to use for speech to text decoding using your Decode script.
The pre-trained model contains the follwing files:
`…
-
This is what I am getting after I run the train command.
(wav2letter) shafkat@server-master:/data/ahnaf/wav2letter/wav2letter/build$ ./Train train --flagsfile=/data/ahnaf/wav2letter/dataset_prep/fl…
-
Hey all,
I'm not convinced by the output of all the CQT transforms for **analysis tasks**, maybe I'm too stupid. I'd like to add more log-freq representations to librosa.
Candidates would be:
- Strai…
-
Hi, I'm trying to use this toolkit for E2E ASR on Chinese, However, after data preparation, I keep getting warnings/errors like:
```
Skipping unknown token '—' when falling back to letter target f…