Speech Synthesis and Emotion

Article Featured Image

BUSINESS PROBLEM: Traditional text-to-speech applications often do not sound very natural, and lack the variations in tone, pitch, speed, etc., that occur during normal human conversations to reflect different emotional states, languages, or settings.

TECHNOLOGY SOLUTION : Speech synthesis and emotion. This technology allows for a more life-like sound to automated voices, making them clear, articulate, expressive, and emotional, depending on the context.

Product: Loquendo TTS
Delivery method: Installed software.
Pricing: Varies, depending on functionality and configurations.
Functionality: Using Loquendo TTS, companies can alter the tone, pitch, speed, or timbre of any voice to express joy, humor, or even anger; and to speak in a formal or informal style, according to context. An Expressive Cues feature provides a series of commonly used expressions said by Loquendo’s voices in a highly expressive and natural way. The application can handle multiple languages with mother-tongue quality, and intelligently insert pauses and vocal intonations that correctly correspond with the narrative of a text. Accuracy of pronunciation has been strengthened with the Morphological Analyser, which examines word structure to resolve pronunciation ambiguities.
Business benefits: Loquendo TTS delivers TTS technology for automated call centers, CRM, news and email-reading, self-service applications, live news, accessing business documents, e-learning, entertainment, automotive telematics, and any embedded application. There’s no need for prerecording, and it can be deployed quickly.
Contact: Loquendo at 39-011-2913403; www.loquendo.com.

Product: FonixTalk 6.0
Delivery method: Downloadable software
Pricing: A one-time licensing fee starts at $2,500, and Fonix receives a royalty on each product sold with its application embedded into it.
Functionality: FonixTalk 6.0 text-to-speech technology supports English, Spanish, German, French, Italian, and Korean, with Chinese and Japanese to be added later this year. It also includes improved females’ and children’s voices and allows developers to enter their own
codes for added voice expressiveness and to alter the speech rate, rhythm, and stress level to reflect different emotional states. Uses of the technology include videogames, handheld dictionaries and translators, PDAs, mobile phones, automotive applications, and assistive technologies.
Business benefits: Developers can create more natural and less robotic-sounding voices, and optimize the sound for specific audio systems, channels, or environments. Improved intonation modeling makes the speech sound more natural. Customized voices can be created in a week or two. The application also does not require additional memory of processor usage, and supports a variety of hardware platforms and operating systems.
Contact: Fonix Speech at 801-553-6600; www.fonix.com.

Product: Nuance RealSpeak
Delivery method: Software that can be installed or embedded.
Pricing: Varies, depending on functionality and configuration.
Functionality: RealSpeak software converts text to speech in more than 26 languages and 40 voices in a variety of accents and speaking styles. RealSpeak can be used to speech-enable an in-car navigation system, provide text message or email reading capabilities for mobile devices, or to enhance call center service. A custom voice program allows users to capture individual voices and have them tailored to their specific needs.
Business benefits: Voices powered by RealSpeak enable organizations to engage with their customers one to one. RealSpeak makes it feasible to use TTS to handle far more system prompts, accelerating development time and reducing overall costs. Organizations can tailor how the application converses with each caller, dynamically modifying prompts and making targeted offers, according to individual circumstances. RealSpeak enhances caller experiences with the seamless transition from spoken prompts to text-tospeech. It supports global organizations with extensive language coverage and extensive dictionaries.
Contact: Nuance at 781-565-5000, www.nuance.com

SpeechTek Covers
for qualified subscribers
Subscribe Now Current Issue Past Issues