Alta Voce SDK - Real-time voice processing
Integrate voice enhancement into your own applications
Voice is everywhere, let it sound.
Our SDK can adapt to your needs for speech-to-speech augmented voice and noise canceling.
Applications include:
- Virtual reality and gaming: sculpt the vocal identity of avatars and characters in real-time
- Voice bots and smart speakers: humanise your synthetic voice by adapting emotions in real time
- Avionics: improve communications between pilots and ground control
- Your imagination is the limit!
Contact us to design the best voice technology for your needs.
Specifications
- C++14 standard (real-time) or Python wrapper (offline)
- Computational cost < 5% CPU
- No dynamic memory allocation
- 35ms processing latency
In addition to the demos below, we also offer a web app where you can try out our effects on your own audio.
Better undestanding
Making speech more intelligible and removing background noise.
Research with our clients shows that voices enhanced with Alta Voce get to engage with more customers than in their normal, unprocessed audio conditions.
Try it in Python
import os
os.environ["VOIMOOO_LICENSE_ID"] = "AAAA-BBBB-CCCC-DDDD"
import pyvoimooo
import pysndfile
wav, fs, enc = pysndfile.sndio.read("yourfile.wav")
syn = pyvoimooo.pvoccombo(
wav,
fs,
efx_inteligibility_scaling=0.5
)
pyvoimooo.writewav("yourfile_intellegible.wav", fs, syn)
Contact us to get a license key
Enhanced smile
Enhancing a speaker’s attitude, e.g. how smiling and welcoming they sound.
Research in the lab shows that smiling voices engage more attention from listeners, triggers spontaneous smiles, and is especially effective on younger listeners.
Try it in Python
import os
os.environ["VOIMOOO_LICENSE_ID"] = "AAAA-BBBB-CCCC-DDDD"
import pyvoimooo
import pysndfile
wav, fs, enc = pysndfile.sndio.read("yourfile.wav")
syn = pyvoimooo.pvoccombo(
wav,
fs,
efx_smile_alpha=1.2
)
pyvoimooo.writewav("yourfile_smiley.wav", fs, syn)
Contact us to get a license key
Shaping voice gender
What if you could shape in real time the way your voice gender is perceived in a call?
English female voice
Less female voice gender
Even less female voice gender
Male voice gender
Try it in Python
import os
os.environ["VOIMOOO_LICENSE_ID"] = "AAAA-BBBB-CCCC-DDDD"
import pyvoimooo
import pysndfile
wav, fs, enc = pysndfile.sndio.read("yourfile.wav")
syn = pyvoimooo.pvoccombo(
wav,
fs,
pitch_scaling=1.5,
frequency_warping=2,
)
pyvoimooo.writewav("yourfile_female.wav", fs, syn)
Contact us to get a license key
It's how you say it
Varying expressivity
Martin (French)
Martin more dynamic
Martin even more dynamic
Our technology can gradually shape the expressivity in a voice, e.g. to make it more dynamic.
Varying prosody inflections
Federica (TTS, Italian)
Buongiorno 1
Buongiorno 2
Buongiorno 3
It is common knowledge that how we say something is more important than what we say. Our research proved it, and our technology allows to find the expressive features conveying the desired intention. For example, listen to Federica’s voice more trustworthy.
Federica (TTS, italian)
Federica trust