I am looking for a better alternative to the 'Alice' voice provided by Twilio. I am pretty sure Twilio only provides two basic default voices along with 'Alice', a more robust version able to more effectively enunciate text. The only problem is that 'Alice' does not sound as natural as other voices used by other services known to be using Twilio. Does anyone have a suggestion as to how to access a better voice? The messages in the call flows will be somewhat dynamic so I don't think using recordings would be practical.
Thanks!
Alice is our most advanced voice that supports additional languages and locales.
Some things you can try to make Alice work better for you might include looking at the <Pause> verb for more deliberate separation between sentences.
If you'd like to get additional control as it sounds like you have experienced with other services using Twilio, you could consider prerecording the static parts and deliver them using <Play> and only using <Say> for your dynamic content, though I know this isn't ideal.
There are some additional hints here https://www.twilio.com/docs/api/twiml/say#hints.
I hope this helps!
Related
I am working in an application that gathers a user's voice input for an IVR. The input we're capturing is a limited set of proper nouns but even though we have added hints for all of the possible options, we very frequently get back unintelligible results, possibly as a result of our users having various accents from all parts of the world. I'm looking for a way to further improve the speech recognition results beyond just using hints. The available Google adaptive classes will not be useful, as there are none that match the type of input that we're gathering. I see that Twilio recently added something called experimental_utterances that may help but I'm finding little technical documentation on what it does or how to implement.
Any guidance on how to improve our speech recognition results?
Google does a decent job doing recognition of proper names, but not in real time just asynchronously. I've not seen a PaaS tool that can do this in real time. I recommend you change your approach and maybe identify callers based on ANI or account number or have them record their name for manual transcription.
david
I am using Twilio.com for telephony, in the U.S.
I have an existing CONFERENCE, and I want to add arbitrary pre-recorded audio to it, at certain times. I am aware of the "call out to yourself from yourself, then use PLAY or SAY" technique.
Is having another "call" going during the (at least two calls) conference really the best/cheapest way to accomplish this?
Twilio developer evangelist here.
That is the best and only way to achieve that feature right now.
However, that is going to change. Recently we added a feature called Agent Conference, which gives more power to the agent and supervisor within a contact centre. While it doesn't help now, the announcement that the initial features for Agent Conference were out of beta also mentioned some future features that are coming, including:
Conference announcements using <Say> or <Play> to everyone in a Conference or to individual participants
Keep an eye on the Twilio blog for further announcements.
I am now doing an NLP project which needs some resources from twitter.
I want to get those tweets posted by "real people" instead of any kind of "official accounts", including celebrities, ads, institutions, media, etc. such as #CNN #TodayWeather #obama #DailySale #BestPrice #FashionTrend.
Hence, is there a better way to do so?
I have considered about it for a long time. By using twitter's API, the returned JSON includes a key called "verified". This can be used to detect weather an account is that kind of "official account". However, today, this blue "V" tick is not only for those shining celebrities. Anyone can apply for it as long as they are a real person. So, I think using this solution will rule out a lot of precious resources.
Moreover, I also considered using textual spam filter. yeah, of course, they are quite good in most cases. However, some accounts, such as #FT, their posts never sound like a spammy ad. But it is not what I want.
I want to ask for a better solution. It can be a long term solution, such as using NLP and NeuroNets to learn from labels. But, well, a prompt solution will be very welcomed.
THX
In the Twilio platform, you can create an "Application" to bundle common configuration details for phone numbers. From https://www.twilio.com/docs/api/rest/applications:
An application inside of Twilio is just a set of URLs and other configuration data that tells Twilio how to behave when one of your Twilio numbers receives a call or SMS message.
Is there a limit to how many phone numbers can be joined into a single Application?
Twilio developer evangelist here.
"TwiML Applications" like the one you mention are designed to be aliases easy to use on a ton of numbers at once, we have a lot of very high volume customers using them on 10,000+ Twilio numbers and probably even a lot more.
In case you need to go beyond those numbers, it's always worth contacting sales as they will be able to increase that for you accordingly while making sure you scale well.
Hope this helps you.
I'm currently working on an automated voice call interpreter. I know that it's not currently possible to gather digits during a conference call directly. That's why I had to come up with a workaround and let twilio dial into the conference call, so that I can gather the participants digits. It works great but the problem is that it's costing me around 4.2c per minute, which of course is not ideal at all.
Is this feature currently on your to do list and if yes, when are you going to make it publicly available?
Cheers,
Marcel
Twilio evangelist here.
Depending on what you want to do this technique might get you better results. It leverages the hanguponstar attribute of the <Dial> verb:
Twilio call transfer from in-call
Hope that helps.