It seems that the announcement and the main page for the repository fail to mention the language that this was tested on (English).
Maybe good to specify? #BenderRule
-
-
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
- Još 3 druga odgovora
Novi razgovor -
-
-
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
Very good results, congrats! Why do we call those models wave to **** in ASR? We don't really use waveforms but log mel-scale filterbanks. "Wave" to me is the raw 1-dimensional input of the audio wave.
-
The wav2letter project has several ASR pipelines, some of which operate on raw waveforms. But the systems in this particular release don't.
Kraj razgovora
Novi razgovor -
-
-
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
Note that we
@merl_news put out a paper last week with SOTA 7.0 % WER on LibriSpeech test-other (vs wav2letter@anywhere's 7.5%) with 590 ms theoretical latency using joint CTC-Transformer with PTDLSTM & triggered attention. Check it out: https://arxiv.org/abs/2001.02674Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
Why was it removed?
-
All of facebook AI's blos posts seem to be down right now
- Još 2 druga odgovora
Novi razgovor -
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.