The 2015 Speech Industry Star Performers: Acapela
Acapela Delivers Authentic Voices for All
Acapela Group has never shied away from the tough challenges with speech technologies. Whether it's providing text-to-speech (TTS) voices for nontraditional languages—such as North Sami, which is spoken by only about 20,000 people in northern Scandinavia—or providing a voice to patients with Lou Gehrig's disease, Acapela was front and center this year.
The Belgium-based company has a unique approach to languages and audio cultural heritage and a dedication to creating voices for all, according to CEO Lars-Erik Larsson. That was a driving force behind its December launch of the My-Own-Voice service, enabling users diagnosed with speech or language disorders to capture their own voice before losing it. This invention will enable users to create TTS applications with their own voice; or they can have another person, such as a close family member, do the recording, in effect donating her voice. Either way, users can avoid having to use standard anonymous, synthetic voices.
Deborah Dahl, principal at Conversational Technologies, has been "very impressed" by the My-Own-Voice effort. This project, she says, "is aimed at making it possible for people who are unable to speak to have their own custom TTS voices."
Larsson says that the launch was just one step in the company's accessibility efforts. Acapela has spent the months since backing that up. In January, the company introduced Acapela TTS voices for Non-Visual Desktop Access (NVDA), enabling blind and visually impaired people to use Acapela’s range of TTS voices with the NVDA screen reader.
And because children use TTS too, Acapela worked with AssistiveWare to launch Infovox iVox 3.1, which brings four children's voices to Mac OS X. These voices, which include boy and girl voices for U.S. and U.K. English, work with any Mac application, including Apple VoiceOver, TextEdit, Preview, Safari, Adobe Acrobat Reader, AssistiveWare’s VisioVoice, Proloquo, and GhostReader. The voices were recorded by and for children.
Acapela is also working to bring speech technology to the masses and in February upgraded its Acapela-Box.com, an online service that enables anyone to produce voice files. This solution for converting text to speech is helping to voice-empower e-learning, videos, presentations, answering machines, blogs, Web sites, interactive voice response systems, animation, MP3 players, and more. In 2014, customers from more than 116 countries used the solution to convert more than 5 million characters into speech.
For companies looking to do the same, Acapela in March released version 2.8 of its Virtual Speaker, which instantly converts text into audio using a number of languages, voices, and output file formats. The product's interface enables users to add tags and sounds, switch voices or languages, and select the most suitable vocalization from a set of alternative pronunciations. Right within the app, users can fine-tune text and pronunciations and highlight text they're currently reading. Acapela also added Korean to the roster of languages and 19 new voices to the Virtual Speaker repertoire.
And the company is active in bringing speech technologies to other markets, including public transportation. In September, the company launched the Acapela Transport business unit, specifically dedicated to meet growing demands for voice solutions adapted to public transportation. Acapela's TTS solutions will be used to create real-time public announcements in 30 languages and more than 100 voices. Custom voices can also be created.
Another area of innovation for Acapela: the gaming arena, which Larsson says "has not yet entered the speech dimension." To help take it there, the company in January teamed with several other European firms on a project that could ultimately make it possible for game players speaking different languages to interface using the same voice. D-Box, as the project is being called, seeks to develop an architecture for conversational agents to support typewritten and/or spoken collaboration in the users' native languages.
"Acapela's focus on text-to-speech synthesis has led to a wide variety of voices in many languages, providing high-quality choices for developers," says Bill Meisel, president of TMA Associates and executive director of the Applied Voice Input/Output Society (AVIOS).
Dahl agrees. "Acapela Group is notable for the large number of TTS voices, languages, and platforms they cover," she says.
AssistiveWare has teamed with Acapela Group to create two children's text-to-speech voices in Swedish.