Manish

RNNT using Lingvo framework

1

Hi @HawkAaron Thanks for this implementation. I found it very usefull. Do you have any plans to implement this using Lingvo framework?

RNN-T for ASR task

1

I want to implement RNN-T for asr. I am finding it difficult to start implementation. Is it even a good idea to use Lingvo to implement RNN-T for streaming ASR?...

ASR model with unidirectional lstm and Monotonic attention not converging

4

I made some changes to asr model. encoder -> unidirectional lstm and monotonic attention. After around 10000 steps, I can see that the loss is around 0.1 ``` I0106 06:35:36.474763...

Hi, I am implementing a model in Lingvo and getting OOM with a tensor of shape [12, 620, 50, 1024]. I am using a 8 GPU (Tesla K80, 8 GB...

Hi, I am implementing an ASR model and getting error in tf.broadcast_to Below is the part of code in ComputePredictions ``` activation = lm_out.last_hidden encoder_outputs = tf.transpose(encoder_outputs, [1, 2, 0])...

Online Decoding

1

Hi, I have trained a model with unidirectional lstm and local attention. How can I use this model for online decoding? Where should I make changes in the existing code?...

Convert Returnn model to TFLite for on-device deployment

2

I want to convert the [returnn full attention setup model](https://github.com/rwth-i6/returnn-experiments/tree/master/2018-asr-attention/librispeech/full-setup-attention) to TFLite for on-device deployment. Should I reimplement the model in TF2.0, load weights and convert to TFLite? Or any...