asr-server
asr-server copied to clipboard
FastCGI support for Kaldi ASR
~/apiai/asr-server$ make make -C src make[1]: Entering directory '/home/osboxes/apiai/asr-server/src' ar -cr libstidecoder.a Timing.o Response.o RequestRawReader.o ResponseJsonWriter.o ResponseMultipartJsonWriter.o OnlineDecoder.o Nnet3LatgenFasterDecoder.o QueryStringParser.o FcgiDecodingApp.o ranlib libstidecoder.a g++ -shared -o liblibstidecoder.so -Wl,--no-undefined -Wl,--as-needed -Wl,-soname=liblibstidecoder.so,--whole-archive...
I have followed the steps given However i always get the following output from the asr server {"status":"ok","data":[{"confidence":0.862751,"text":""}],"interrupted":"endofspeech","time":1080} Please guide on how to check the asr logs
Hey , thanks for the server as well as for pre-trained model . i am kind of new at kaldi and server management , can anyone guide me how to...
I'm not using `ivector` for nnet3 model, referring to the help: ``` --online : You can set this to false to disable online iVector estimation and have all the data...
ERROR (fcgi-nnet3-decoder[5.5.268~1-f9828]:DecodableNnetLoopedOnlineBase():decodable-online-looped.cc:50) Ivector feature dimension mismatch: got -1 but network expects 100 [ Stack-Trace: ] kaldi::MessageLogger::LogMessage() const kaldi::MessageLogger::LogAndThrow::operator=(kaldi::MessageLogger const&) kaldi::nnet3::DecodableNnetLoopedOnlineBase::DecodableNnetLoopedOnlineBase(kaldi::nnet3::DecodableNnetSimpleLoopedInfo const&, kaldi::OnlineFeatureInterface*, kaldi::OnlineFeatureInterface*) kaldi::SingleUtteranceNnet3DecoderTpl::SingleUtteranceNnet3DecoderTpl(kaldi::LatticeFasterDecoderConfig const&, kaldi::TransitionModel const&, kaldi::nnet3::DecodableNnetSimpleLoopedInfo const&, fst::Fst...
How am I supposed to refer to the spawned fcgi with curl exactly ? The curl command doesn't even specify the port, and definitely something is not working as curl...
I followed all the steps. apache2 is correctly installed after running **/apiai/asr-server/fcgi-nnet3-decoder --fcgi-socket=:8000** m getting the following output on terminal /apiai/asr-server/fcgi-nnet3-decoder --feature-type=mfcc --mfcc-config=mfcc.conf --frame-subsampling-factor=3 --max-active=2000 --beam=15.0 --lattice-beam=6.0 --acoustic-scale=1.0 --endpoint.silence-phones=1 --endpoint.rule1.min-trailing-silence=0.5...
Hi, What type of server config would it need to process/decode 10 concurrent speech recognition? How many cores and ram? Not training only for decoding. Thanks
``` ERROR (fcgi-nnet3-decoder[5.2.62~1-a2342]:ToInt():parse-options.cc:598) Invalid integer option "" [ Stack-Trace: ] ../asr-server/fcgi-nnet3-decoder() [0xf2e23a] kaldi::MessageLogger::HandleMessage(kaldi::LogMessageEnvelope const&, char const*) kaldi::MessageLogger::~MessageLogger() kaldi::ParseOptions::ToInt(std::__cxx11::basic_string const&) kaldi::ParseOptions::SetOption(std::__cxx11::basic_string const&, std::__cxx11::basic_string const&, bool) kaldi::ParseOptions::Read(int, char const* const*) apiai::FcgiDecodingApp::Run(int, char**)...
The answer is always like this.How could I fix it? 