Search Results for "2-plan"

...Mimi processes 24 kHz audio, down to a 12.5 Hz representation with a bandwidth of 1.1 kbps, in a fully streaming manner (latency of 80ms, the frame size), yet performs better than existing, non-streaming, codecs like SpeechTokenizer (50 Hz, 4kbps), or SemantiCodec (50 Hz, 1.3kbps). Moshi models two streams of audio: one corresponds to Moshi, and the other one to the user. At inference, the stream from the user is taken from the audio input, and the one for Moshi is sampled from the model's output. Along these two audio streams, Moshi predicts text tokens corresponding to its own speech, its inner monologue, which greatly improves the quality of its generation. ...

Choose from the "voices" available, and type in what you want the computer to say. A wave file called sounds.wav is stored to the output sub folder. Output is intended primarily for users who need speech for animated characters in videos.

HMM Speech Recognition in Java

This project provide hidden Markov model speech recognition system by using Matlab/Simulink/Stateflow.

...Thiruvalluvar, the Tamil Sage of the 1st Century CE had said: “Wealth of wealth is wealth acquired be ear attent; Wealth mid all wealth supremely excellent. “ Kural No : 411 This software is the first version, with limited words in Tamil for them to practice. We have tested it with the help of a school and atleast two children were able to pick up some words, with just a little practice. It is our hope that with more words and a tablet pc, they can help themselves to practice continuously and attain a level of proficiency that would make life easier on them. We are releasing this software now, as we hope that it would encourage more talented developers and designers to put in their expertise and make this a workable idea with out any language and regional barriers, so that it can be useful to all the 416 million disabled without any cost.

An initiative to create something similar to the windows program Roger Wilco, Teamspeak, BattleCom and Speak Freely, allowing users from different platforms talk with each other in real time with minimal CPU and bandwidth usage. Voice chat.....

...This support includes Croatian lexicon, which contains 83 entries, Croatian synthesis module, which contains Croatian phoneset, lts rules, tokenization, utterance, and accents, Croatian support for mbrola speech synthesizer, and 2 Czech voices provided by brailcom, until completing Croatian festival voice.

This project is being developed to be a Java based speech recognition (SR) program. In addition to the SR program itself, it includes a program which allows a user to view the sound being received by the computer. The user can manipulate this data.

Cairo sets out to provide an enterprise grade, MRCPv2 compliant speech solution utilizing existing open source speech resources such as FreeTTS and Sphinx-4.

Speech Made Visible is an experiment in showing some of the qualities of speech in printed text. Analyze a recording for attributes like pitch, intensity (loudness), and speed; then style the words in a transcript to suggest those characteristics.

VALARI is a Virtual Announced Launch And Recovery Informant for Mac OSX for GPS & Telemetry tracking of modern high power rocketry. Eventually we might port this to other OSes.

Project that aims converting a text page directly into MP3 or other audio format using the MBrola libraries

audacity-extra now provides a sleek dark themed version of the Audacity open source sound editor. The project experiments with Audacity variations. There's a vowel-sound target-practice display for language learners and an analog waveform data logger for embedded systems.

Eve is a AI project written in python that takes commands verbally or textually to control the computer and eveyday functions.

A text to speech converter which will be able to read any document(Presently it is reading text and .doc files).The main aim of the project is to make reading an interesting task and assist BLIND people.

Auvai is a Java API and Java Swing based application for Text to Speech conversion of Unicode Tamil. Future direction of this API and application is to support Text to Speech conversion for all "Indic" languages.

GibPhone is a highly extensible VoIP/IM client for the .NET framework that uses a powerful plugin engine to allow for UI extensions and any call control stack / media payload / transport protocol combination.

Cowpie is an application for synchronizing character actions with sound files and animation tools like Blender and 3DS Max. Such actions include the phonemes of speech and facial expressions. Cowpie will also coordinate speech for multiple characters.