Please provide the following information when requesting support.
kindly follow this post:
Followed the exact same steps, but getting error as language - ar is not recognized.
riva build command:
riva-build speech_recognition /data/Conformer-CTC-L_spe128_ar-AR_3.0.rmir /data/Conformer-CTC-L_spe128_ar-AR_3.0.riva --language_code=ar-AR --decoder_type=nemo
Hardware - GPU (A6000)
Hardware - CPU
Operating System
Riva Version - 2.18
TLT Version (if relevant)
How to reproduce the issue ? (This is for errors. Please share the command and the detailed log here)
import io
import librosa
from time import time
import numpy as np
import IPython.display as ipd
import grpc
import requests
# NLP proto
import riva_api.riva_nlp_pb2 as rnlp
import riva_api.riva_nlp_pb2_grpc as rnlp_srv
# ASR proto
import riva_api.riva_asr_pb2 as rasr
import riva_api.riva_asr_pb2_grpc as rasr_srv
# TTS proto
import riva_api.riva_tts_pb2 as rtts
import riva_api.riva_tts_pb2_grpc as rtts_srv
import riva_api.riva_audio_pb2 as ra
channel = grpc.insecure_channel('localhost:50051')
riva_asr = rasr_srv.RivaSpeechRecognitionStub(channel)
riva_nlp = rnlp_srv.RivaLanguageUnderstandingStub(channel)
riva_tts = rtts_srv.RivaSpeechSynthesisStub(channel)
# This example uses a .wav file with LINEAR_PCM encoding.
# read in an audio file from local disk
#path = "/home/ubuntu/data2/riva_/en-US_sample.wav"
path = "/home/ubuntu/data2/riva_/ar-AR_sample.wav"
audio, sr = librosa.core.load(path, sr=None)
with io.open(path, 'rb') as fh:
content = fh.read()
ipd.Audio(path)
req = rasr.RecognizeRequest()
req.audio = content # raw bytes
req.config.encoding = ra.AudioEncoding.LINEAR_PCM # Supports LINEAR_PCM, FLAC, MULAW and ALAW audio encodings
req.config.sample_rate_hertz = sr # Audio will be resampled if necessary
#req.config.language_code = "en-US" # Ignored, will route to correct model in future release
req.config.language_code = "ar-AR" # Ignored, will route to correct model in future release
req.config.max_alternatives = 1 # How many top-N hypotheses to return
req.config.enable_automatic_punctuation = True # Add punctuation when end of VAD detected
req.config.audio_channel_count = 1 # Mono channel
response = riva_asr.Recognize(req)
asr_best_transcript = response.results[0].alternatives[0].transcript
print("ASR Transcript:", asr_best_transcript)
print("\n\nFull Response Message:")
print(response)
and Error Code is
Traceback (most recent call last):
File "/home/ubuntu/data2/riva_/eval.py", line 48, in <module>
response = riva_asr.Recognize(req)
^^^^^^^^^^^^^^^^^^^^^^^
File "/home/ubuntu/miniconda3/lib/python3.12/site-packages/grpc/_channel.py", line 1181, in __call__
return _end_unary_response_blocking(state, call, False, None)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/ubuntu/miniconda3/lib/python3.12/site-packages/grpc/_channel.py", line 1006, in _end_unary_response_blocking
raise _InactiveRpcError(state) # pytype: disable=not-instantiable
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
grpc._channel._InactiveRpcError: <_InactiveRpcError of RPC that terminated with:
status = StatusCode.INVALID_ARGUMENT
details = "Error: Unavailable model requested given these parameters: language_code=ar; sample_rate=16000; type=offline; "
debug_error_string = "UNKNOWN:Error received from peer {grpc_message:"Error: Unavailable model requested given these parameters: language_code=ar; sample_rate=16000; type=offline; ", grpc_status:3, created_time:"2025-02-25T11:01:46.524261779+00:00"}"
>
config.sh script file changes:
diff config.sh config.sh.bak
47c47
< asr_language_code=("ar-AR")
---
> asr_language_code=("en-US")
72c72
< tts_language_code=("ar-AR")
---
> tts_language_code=("en-US")
104,106c104
< # riva_model_loc="/home/ubuntu/data2/riva_/dck_mount"
< #riva_model_loc="/home/ubuntu/data2/riva_/riva_quickstart_v2.18.0/arabic_rmir"
< riva_model_loc="/home/ubuntu/data2/riva_/riva_quickstart_v2.18.0/baseline_name"
---
docker logs:
docker_riva_logs.txt (19.8 KB)
I am able to probe the model, seems to me ; language-code issue, kindly help further to fix this issue and progress with my work further.