-
File ~/work/python/tts/MeloTTS/melo/api.py:81, in TTS.tts_to_file(self, text, speaker_id, output_path, sdp_ratio, noise_scale, noise_scale_w, speed)
79 t = re.sub(r'([a-z])([A-Z])', r'\1 \2'…
-
Cohere's new Command-R-Plus model reportedly features a 128k context window. However, testing with progressively longer prompts reveals it begins producing nonsensical output (e.g., "\\...") after 819…
-
The multilingual models were trained on both speech recognition and speech translation. The function of free translation into different languages would be very much in demand, given that besides Googl…
subof updated
3 months ago
-
Reduce usage of [Immutable](https://immutable-js.github.io/immutable-js/) is meaningful.
Immutablity is good, but [Immutable.js](https://immutable-js.github.io/immutable-js/) is hard.
- [Immutab…
-
Pi is capable of generating text responses in non-English languages. However, when Pi reads those text responses aloud, it speaks with an English accent, whatever the language of the text.
This app…
-
**Describe the bug**
I followed the example of bedrock https://github.com/VinciGit00/Scrapegraph-ai/blob/main/examples/bedrock/smart_scraper_bedrock.py
It was working in the first place. Then af…
-
### Describe the bug
Here is the bug:
```
PS E:\AI> tts --model_name "tts_models/multilingual/multi-dataset/xtts_v1" --text "Ceci est un teste de voix." --language_idx "fr" --use_cuda False
> tt…
-
I am reaching out for assistance with an issue I'm experiencing while trying to use the intfloat/multilingual-e5-large model in a TypeScript project in my local environment. I am utilizing LangChain.j…
-
Currently human evaluation lists 0.0 both for average and best human rater: https://github.com/google/BIG-bench/tree/main/bigbench/benchmark_tasks/abstraction_and_reasoning_corpus
I wonder if there i…
-
Original comment: https://gitlab.hltcoe.jhu.edu/research/turkle/issues/173#note_28038
Pasting 😃 into a text field for a project results in a MySQL error. Doing this same in a CSV or in a field on a t…