-
When longer audio(up to 40 seconds) is used for inference only for first 5-6 seconds transcripts are generated successfully but rest of the part of the audio has no or one word in the final transcript…
-
```
asr_python:
model: 'conformer_wenetspeech'
lang: 'zh'
sample_rate: 16000
cfg_path: # [optional]
ckpt_path: # [optional]
decode_method: 'attention_rescoring'
num…
-
## What is the current situation?
AFV bridge and RDF plugins are versions included:
- https://github.com/AndyTWF/afv-euroscope-bridge
- https://github.com/chembergj/RDF
## What is the proposed…
-
With the following definitions:
```python
from lpython import dataclass, i64
@dataclass
class TransactionItem:
account_idx: i64
commodity: str
amount: i64
@dataclass
class Tra…
-
### Testmerges
BYOND Version: 515.1610 Server revision compiled on: 2023-08-30T12:16:46+00:00 Master commit: 4ad2587f9a8910c34e476d26305acbc829a4fd20 The following pull requests are currently test me…
-
Dear Steven,
many thanks for providing this awesome package. However, I have some trouble with the results pdf as labels don't align properly (which is actually related to this [issue](https://gith…
-
when I run "self.asr_model.load_weights()"
an error occurred:Read fewer bytes than requested
-
We have been using google ASR.
Code : https://github.com/fonoster/fonoster/blob/main/mods/googleasr/src/tracker.ts
Instead of real time result, latency lies in range of 2s to 5s detoriating user…
-
| CPtrToPointer(expr cptr, expr ptr, expr? shape, expr? lower_bounds)
It converts a C pointer (CPtr ttype) in `cptr` to a Fortran array in `ptr`, one has to prescribe the `shape` (not su…
-
:~/Desktop/espnet/egs2/an4/asr1$ . path.sh
(espnet):~/Desktop/espnet/egs2/an4/asr1$ ./run.sh
....
...
2022-12-07T00:35:45 (asr.sh:809:main) Stage 6: LM collect stats: train_set=dump/raw/lm_tra…