DSPRelated.com
Forums

Speech audio file format, processing, archives and audio file reading codes/libraries

Started by Unknown January 21, 2015
What is the current trend in using the audio file format ... aiff, mp3, wav ... ?

What are the commercial speech recognition systems like Nuance or Google API .... use?

Are there any archives, where one can download recorded speech files?

Finally, any codes (C++ or whatever)/libraries for reading audio files so these can be processed?

Thanks 
On 1/21/2015 12:18 PM, speech2020@gmail.com wrote:
> What is the current trend in using the audio file format ... aiff, mp3, wav ... ?
That would be WAV. It contains the most information and is therefore larger. If file size is a consideration, MP3 is used.
> > What are the commercial speech recognition systems like Nuance or Google API .... use? >
En Ingl�s, por favor?
> Are there any archives, where one can download recorded speech files?
http://rachelsenglish.com/video-category/sounds https://www.internationalphoneticassociation.org/content/sound-recordings Google TIMIT.
> > Finally, any codes (C++ or whatever)/libraries for reading audio files so these can be processed?
Sphinx -rewritten in Java- has become the de facto leader. So much, that other competing projects have shut down. They have some stuff in C, have not seen anything in C++. http://www.speech.cs.cmu.edu/ -Ramon