-
Notifications
You must be signed in to change notification settings - Fork 3.1k
how to run specific Riva ASR model #6868
Replies: 2 comments · 4 replies
-
@VahidooX , could you take a look at this? |
Beta Was this translation helpful? Give feedback.
All reactions
-
👍 1
-
This specific tutorial is old and just works with models based on Jasper/Quartzent. For Conformer models you may not use it. If you want to try a real and more efficient streaming you may use Riva. |
Beta Was this translation helpful? Give feedback.
All reactions
-
👍 1
-
@VahidooX Thanks a lot for your useful reply, could you please tell me more about using Riva for more efficient streaming as you mentioned,what do you mean by that ,but please bear with me as I'm new to this area. |
Beta Was this translation helpful? Give feedback.
All reactions
-
There are some dev examples here - https://developer.nvidia.com/riva It is a full deployment framework with efficient streaming inference. If you only want offline inference you can try Nemo itself. You can use Nemo for inefficient streaming inference with the scripts provided above but we don't have that in notebook form |
Beta Was this translation helpful? Give feedback.
All reactions
-
👍 1
-
Actually i need efficient streaming inference, i will try Riva as you said and update you, thanks a lot all of you for your great help 👍 |
Beta Was this translation helpful? Give feedback.
All reactions
-
I tried to use rive to deploy the model, i converted the acoustic model and lm to the desired formats and I made the pipeline, but I got an error when trying to start the Riva server as below : Hardware - GPU (NVIDIA GeForce RTX 3050 Ti Laptop GPU) i tried to start riva speeech by typing this command in git bash (./riva_start.sh config.sh) but got this error, please help me to solve it: moham@Mohamed_Salama MINGW64 ~/Desktop/Nvidia_course/riva_quickstart_v2.3.0
$ ./riva_start.sh config.sh
Starting Riva Speech Services. This may take several minutes depending on the number of models deployed.
9e366e5c90a2c640d777343a946e978f23ba576279d3a700d2485ec2490d670e
OCI runtime exec failed: exec failed: unable to start container process: exec: "C:/Program Files/Git/usr/bin/grpc_health_probe": stat C:/Program Files/Git/usr/bin/grpc_health_probe: no such file or directory: unknown
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Waiting for Riva server to load all models...retrying in 10 seconds
Health ready check failed.
Check Riva logs with: docker logs riva-speech docker logs:
|
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
Hi,
I'm new to the field of ASR, and I saw one great STT model called "Conformer-CTC-L_spe128_ar-AR_3.0.nemo" which i downloaded from your website (link)
I tried to apply the code in your notebook called (Online_ASR_Microphone_Demo.ipynb) link
it works fine when using the same model in the notebook ('QuartzNet15x5Base-En')
using this code :
asr_model = nemo_asr.models.EncDecCTCModel.from_pretrained('QuartzNet15x5Base-En')
when i replaced the model with the other model "Conformer-CTC-L_spe128_ar-AR_3.0.nemo" (link) and i used it using this code:
asr_model = nemo_asr.models.ASRModel.restore_from(restore_path="Conformer-CTC-L_spe128_ar-AR_3.0.nemo")
the cell called (Streaming Inference) keep showing errors
cell:
error:
please help me to solve this issue, how i can use this model, is that the best way, or is there is another way?
Thanks.
Beta Was this translation helpful? Give feedback.
All reactions