



## vosk

- <https://github.com/alphacep/vosk-api>
- <https://github.com/alphacep/vosk-server>

### Vosk Server

Server: requires about `8G` memory or more.

```
docker run -d -p 2700:2700 alphacep/kaldi-en:latest
```

Client
```
git clone https://github.com/alphacep/vosk-server
cd vosk-server/websocket
./test.py test.wav
```

Client with microphone
```
pip3 install sounddevice
./test_microphone.py -u ws://localhost:2700
```

## Julius

<https://github.com/julius-speech/julius>

### Build

```
git clone https://github.com/julius-speech/julius.git
cd julius
```

Fix for macOS: Add belowto `libsent/src/adin_mic_darwin_coreaudio.c`
```c
#include <sent/stddefs.h>
```

```
./configure --enable-words-int
make -j4
```

### Run

Download [JuliusModel](https://sourceforge.net/projects/juliusmodels/files/) named `ENVR-v5.4.Dnn.Bin.zip`

```
unzip ENVR-v5.4.Dnn.Bin.zip
cd ENVR-v5.4.Dnn.Bin
```

Edit config file `dnn.jconf`

```
-feature_options -htkconf wav_config -cvn -cmnload ENVR-v5.3.norm -cmnstatic
+feature_options -htkconf wav_config -cvn -cmnload ENVR-v5.3.norm -cvnstatic

+state_prior_log10nize false
```

Recognize audio file

```
../julius/julius/julius -C julius.jconf -dnnconf dnn.jconf
```

Edit `test.dbl` to test with other audio files.


<h2>vosk</h2>
<ul>
<li><a href="https://github.com/alphacep/vosk-api" target="_blank">https://github.com/alphacep/vosk-api</a></li>
<li><a href="https://github.com/alphacep/vosk-server" target="_blank">https://github.com/alphacep/vosk-server</a></li>
</ul>
<h3>Vosk Server</h3>
<p>Server: requires about <code>8G</code> memory or more.</p>
<pre class="language-bash"><code>docker run -d -p <span class="token number">2700</span>:2700 alphacep/kaldi-en:latest
</code></pre>
<p>Client</p>
<pre class="language-bash"><code><span class="token function">git</span> clone https://github.com/alphacep/vosk-server
<span class="token builtin class-name">cd</span> vosk-server/websocket
./test.py test.wav
</code></pre>
<p>Client with microphone</p>
<pre class="language-bash"><code>pip3 <span class="token function">install</span> sounddevice
./test_microphone.py -u ws://localhost:2700
</code></pre>
<h2>Julius</h2>
<p><a href="https://github.com/julius-speech/julius" target="_blank">https://github.com/julius-speech/julius</a></p>
<h3>Build</h3>
<pre class="language-bash"><code><span class="token function">git</span> clone https://github.com/julius-speech/julius.git
<span class="token builtin class-name">cd</span> julius
</code></pre>
<p>Fix for macOS: Add belowto <code>libsent/src/adin_mic_darwin_coreaudio.c</code></p>
<pre class="language-c"><code class="language-c"><span class="token macro property">#<span class="token directive keyword">include</span> <span class="token string">&#x3C;sent/stddefs.h></span></span>
</code></pre>
<pre class="language-bash"><code>./configure --enable-words-int
<span class="token function">make</span> -j4
</code></pre>
<h3>Run</h3>
<p>Download <a href="https://sourceforge.net/projects/juliusmodels/files/" target="_blank">JuliusModel</a> named <code>ENVR-v5.4.Dnn.Bin.zip</code></p>
<pre class="language-bash"><code><span class="token function">unzip</span> ENVR-v5.4.Dnn.Bin.zip
<span class="token builtin class-name">cd</span> ENVR-v5.4.Dnn.Bin
</code></pre>
<p>Edit config file <code>dnn.jconf</code></p>
<pre class="language-bash"><code>-feature_options -htkconf wav_config -cvn -cmnload ENVR-v5.3.norm -cmnstatic
+feature_options -htkconf wav_config -cvn -cmnload ENVR-v5.3.norm -cvnstatic

+state_prior_log10nize <span class="token boolean">false</span>
</code></pre>
<p>Recognize audio file</p>
<pre class="language-bash"><code><span class="token punctuation">..</span>/julius/julius/julius -C julius.jconf -dnnconf dnn.jconf
</code></pre>
<p>Edit <code>test.dbl</code> to test with other audio files.</p>

CL's Blogs

Speech to text

vosk

Vosk Server

Julius

Build

Run