Recherche avancée

Sur d’autres sites (9517)

• Bluemix node js build pack to support webm audio conversion

  31 août 2018, par Nimmy Mohandas

  Even after adding ffmpeg to bluemix node js build pack(I tried this https://github.com/BlueChasm/nodejs-buildpack-ffmpeg), it doesn’t support webm audio format conversion.Could anyone please suggest alternate ways to support this issue ?

  Description : I want to do speech to text conversion using Google speech recognition. I am using react native for front-end development. I recorded the voice and generated a webm audio file. Since the google speech recognition doesn’t support webm file format, I converted it into wav format using fluent-ffmpeg. It is working perfectly in my local system, but when I deploy it to the IBM cloud foundry using ffmpeg buildpack(https://github.com/BlueChasm/nodejs-buildpack-ffmpeg), it generates a file without extension. Following are my code,

  router.post('/st', audioUpload.single('audio'), function(req, res, next) {
  
          if (!req.file) {
  
              return res.json({ status: false, error: 'No input given!' });
  
          }
  
  
  
          /**
  
           * convert to .wav
  
           */
  
  
  
  
  
          ffmpeg(req.file.path)
  
              .toFormat('wav')
  
              .inputOptions(['-r 32000',
  
                  '-ac 1'
  
              ])
  
  
  
              .on('error', (err) => {
  
                  console.log('An error occurred: ' + err.message);
  
              })
  
              .save(path.join(__dirname, '..', 'uploads/audio/file.wav'))
  
  
  
          .on('end', () => {
  
  
  
              st.detectAudioFile(path.join(__dirname, '..', 'uploads/audio/file.wav')).then(data => {
  
                      fs.unlink(path.join(__dirname, '..', 'uploads/audio/file.wav'), function(err) {
  
                          if (err) return res.send(err);
  
                          const response = data[0];
  
                          const transcription = response.results
  
                              .map(result => result.alternatives[0].transcript)
  
                              .join('\n');
  
                          return res.json({ transcription });
  
                      }), fs.unlink(req.file.path, function(err) {
  
                          if (err) return res.json(err);
  
                      })
  
  
  
                  })
  
                  .catch(err => {
  
                      res.send({
  
  
  
                          status: false,
  
                          error_code: 400,
  
                          err: err.error || err.message
  
                      });
  
                  });
  
          })
  
      })

  //speech-to-text.js

  const fs = require('fs');
  
  const speech = require('@google-cloud/speech');
  
  
  
  
  
  var detectAudioFile = function(fileName) {
  
  
  
      const Speechclient = new speech.SpeechClient({})
  
  
  
      // Reads a local audio file and converts it to base64
  
      const file = fs.readFileSync(fileName);
  
  
  
      const audioBytes = file.toString('base64');
  
      const audio = {
  
          content: audioBytes,
  
      };
  
      const config = {
  
          // encoding: "FLAC",
  
          // sampleRateHertz: 44100,
  
  
  
          languageCode: 'en-US',
  
      };
  
      const request = {
  
          audio: audio,
  
          config: config,
  
      };
  
      return Speechclient
  
          .recognize(request);
  
  
  
  }
  
  
  
  
  
  
  
  module.exports = {
  
      detectAudioFile
  
  }

  `

• lavc : implement an ATRAC9 decoder

  30 juin 2018, par Rostislav Pehlivanov

  lavc : implement an ATRAC9 decoder
  This commit implements a full ATRAC9 decoder, a simple low-delay codec
  
  developed by Sony and used in most PSVita games, some PS3 games and some
  
  PS4 games. Its similar to AAC in that it uses Huffman coded scalefactors
  
  but instead of vector quantization it just Huffman codes the spectral
  
  coefficients (in a way similar to how Opus splits band energy coding
  
  into coarse and fine precision). It opts to write rather large Huffman
  
  codes by packing several small coefficients into one Huffman coded
  
  symbol, though I don't believe this increases efficiency at all.
  
  Band extension implements SBC in a simple way, first it mirrors the
  
  lower spectrum onto the higher frequencies and then it uses one of 5
  
  filters to shape it. Noise substitution is implemented via 2 of them.
  
  Unlike previous ATRAC codecs, there's no QMF, this is a standard MDCT
  
  codec.
  Based off of the reverse engineering work of Alex Barney.
  Signed-off-by : Rostislav Pehlivanov <atomnuker@gmail.com>
  

  • [DH] Changelog
  • [DH] configure
  • [DH] libavcodec/Makefile
  • [DH] libavcodec/allcodecs.c
  • [DH] libavcodec/atrac9dec.c
  • [DH] libavcodec/atrac9tab.h
  • [DH] libavcodec/avcodec.h
  • [DH] libavcodec/codec_desc.c
  • [DH] libavcodec/utils.c
  • [DH] libavcodec/version.h

• Send stream from hdhomerun to AWS EC2 instance

  13 novembre 2018, par icalvete

  I’m trying something odd to get fun. Send stream from hdhomerun to AWS EC2 instance.

  First aproach. Read stream with ffmpeg
  (tested locally successfully)

  hdhomerun_config discover sya my device have ip 192.168.1.200 so I open port on my router in this way

  <my home="home" public="public" ip="ip">:5005 -> 192.168.1.200:80
  
  <my home="home" public="public" ip="ip">:5004 -> 192.168.1.200:5004
  
  </my></my>

  w3m and telnet say everything is ok.

  But...

  from EC2 instance I do

  /usr/bin/ffmpeg -y -i 'http://<my home="home" public="public" ip="ip">:5004/auto/v5057?transcode=internet240' -t 12 -vn -acodec pcm_s16le -ar 16000 -ac 1 '/tmp/test.wav'
  
  </my>

  In tuner1 I can see...

  Virtual Channel none
  
  Frequency   698.000 MHz
  
  Program Number  186
  
  Modulation Lock t8qam64
  
  Signal Strength 89%
  
  Signal Quality  100%
  
  Symbol Quality  100%
  
  Streaming Rate  none
  
  Resource Lock   <ec2 public="public" ip="ip">
  
  </ec2>

  In hdhomerun system logs...

  19700102-10:27:25 Tuner: tuner0 tuning 5057 Telecinco (t8qam64:698MHz-186)
  
  19700102-10:27:25 Tuner: tuner0 streaming http to <ec2 public="public" ip="ip">:34124
  
  </ec2>

  Everything seems ok but ffmpeg don’t get any data.

  Seccond aproach. Send stream to wowza

  I have a wowza server running in EC2 instance

  From a linux box at my home I do.

  root# /usr/bin/hdhomerun_config 1250D7B2 scan /tuner1 scan.log
  
  root# /usr/bin/hdhomerun_config 1250D7B2 set /tuner1/channel auto:651000000
  
  root# /usr/bin/hdhomerun_config 1250D7B2 get /tuner1/streaminfo
  
  root# /usr/bin/hdhomerun_config 1250D7B2 get /tuner1/program 190
  
  root# /usr/bin/hdhomerun_config 1250D7B2 set /tuner1/target rtp://<wowza server="server" public="public" ip="ip">:1935/TEST/hdhr
  
  root# echo $?
  
  root# 0
  
  </wowza>

  Ends without error buy doesn’t work
  This was a long shot but ....

  Any ideas to do this.

  Thanks !!!