Voice to text

Topic Views - 1711

Bright UK


Does anybody have experience with good voice-to-text solutions?

We are looking to translate verbatim comments into text for customer satisfaction surveys, so ultimately looking for a rigid .wav -> .txt solution!

Any comment would be helpful :)



I recently tried to cover this at call centre expo with dissapointing results.

The long and short of it is technology is not yet able to identify changes in pitch and translate them in behaviour (i.e going from happy to angry). It also doesnt recognise grammar, for example, where to put commas and full stops etc. As a proffesional tool vendors werent selling this to me as a service.


I havent used this software before but it may ber what your looking for?


I havent used it so cant advocate it.

Bright UK

Thank you Lee!

No doubt, the accuracy of grammar and pitch will be low. However, I do not think it is of high importance in this case. Just getting the words down into a text format would allow for some interesting data mining given enough surveys (and verbatim).

I was looking at DNS as well, but it mainly seems like a dictating tool for personal use. Nuance seems to have a broad range of services to offer though, including CC tools.

Some research led me to CMU Sphinx, which seems very potent for the right (wizzy) person.

The hunting continues :)


Call Centre Helper

I'm not sure how robust Text-to-Speech solutions are.

It was the downfall for SpinVox (now purchased by Nuance). It was widely suggested that it was manual transcription that was used to convert the messages.


Are you looking to search for words in a call recording. If so there are a number of phonetic based analytics solutions that enable you to search for words. Nexidia and Aurix are two companies that can do this.

Bright UK

Interesting article!

It seems like there are plenty of companies offering the same sort of solution - to manually transcribe audio files.

We are not looking to search for words, rather to translate the audio into text. Apart from analysis, the main reason is to support agents without a sound card who deserve to see verbatim for their own c-sat surveys.

Maybe some of the mentioned companies offer something related, I will have to take a closer look!

Telemarketing Manager


Hi Niklas

You can use the speech to text application on the Avaya modular mesaging platform to do what you want.

It is designed to convert voicemails to text so that people can read them on the move.

Have a look at the link below.

Speech to Text:


I hope this is helpful.

Bright UK

Thanks Jason, definitely gonna contact Avaya!

Bright UK

(2 weeks later) and still no word from Avaya. I have left voice mails and sent emails. Any idea about pricing for such a platform?

Also, would anyone perhaps know of a company with similar solutions in the meantime?


We use NICE Systems Analytics.

Managing Consultant


Hi Niklas,

The Avaya module is probably part of a larger solution and may carry the corresponding price tag (in the same sense, so would most Speech analytics products).

Most R&D into S2T technology is around voicemail conversion and there are quite a few providers (phonetag, jott etc). I would suggest you look at something like Spinvox in the first instance, as you should at least be able to get a reasonable level of transcription from UK recordings.

Hope this helps, Stephen

Bright UK

Thanks Stephen!

Had a chat with VoxSciences and might be able to get somewhere. Their engine is backed up by manual transcribers as well and they do not offer any non-manual service.


Try Rapide Solutions.

[Part of message removed by moderator - please be careful to avoid advertising]

Want to add a comment?

Not found what you were looking for?

1. Try searching through our site.
2. Still not got an answer?

Why not ask the Call Centre Helper Community? Click here to ask your question