TTS Engines Research Articles

Object Recognition System for the Visually Impaired: A Deep Learning Approach using Arabic Annotation

Object detection is an important computer vision technique that has increasingly attracted the attention of researchers in recent years. The literature to date in the field has introduced a range of object detection models. However, these models have largely been English-language-based, and there is only a limited number of published studies that have addressed how object detection can be implemented for the Arabic language. As far as we are aware, the generation of an Arabic text-to-speech engine to utter objects’ names and their positions in images to help Arabic-speaking visually impaired people has not been investigated previously. Therefore, in this study, we propose an object detection and segmentation model based on the Mask R-CNN algorithm that is capable of identifying and locating different objects in images, then uttering their names and positions in Arabic. The proposed model was trained on the Pascal VOC 2007 and 2012 datasets and evaluated on the Pascal VOC 2007 testing set. We believe that this is one of a few studies that uses these datasets to train and test the Mask R-CNN model. The performance of the proposed object detection model was evaluated and compared with previous object detection models in the literature, and the results demonstrated its superiority and ability to achieve an accuracy of 83.9%. Moreover, experiments were conducted to evaluate the performance of the incorporated translator and TTS engines, and the results showed that the proposed model could be effective in helping Arabic-speaking visually impaired people understand the content of digital images.

The First Vietnamese FOSD-Tacotron-2-based Text-to-Speech Model Dataset.

Recent trends in voicebot application development have enabled utilization of both speech-to-text and text-to-speech (TTS) generation techniques. In order to generate a voice response to a given speech, one needs to use a TTS engine. The recently developed TTS engines are shifting towards end-to-end approaches utilizing models such as Tacotron, Tacotron-2, WaveNet, and WaveGlow. The reason is that it enables a TTS service provider to focus on developing training and validating datasets comprising of labelled texts and recorded speeches instead of designing an entirely new model that outperforms the others which is time-consuming and costly. In this context, this work introduces the first Vietnamese FPT Open Speech Data (FOSD)-Tacotron-2-based TTS model dataset. This dataset comprises of a configuration file in *.json format; training and validating text input files (in *.csv format); a 225,000-step checkpoint of the trained model; and several sample generated audios. The published dataset is extremely worth for serving as a model for benchmarking with other newly developed TTS models / engines. In addition, it opens an entirely new TTS research optimization problem to be addressed: How to effectively generate speech from text given: a black box TTS (trained) model and its training and validation input texts.

On the intelligibility of fast synthesized speech for people who are blind: A cross-system comparison.

People who are blind increasingly use synthesized speech as a primary output modality when interacting with computers, mobile devices, and web-based services. They typically prefer to listen to the synthesized speech at speeds multiple times real time, and consequently develop strong preferences for particular text-to-speech (TTS) engines and voices. These usage-based preferences lead to potentially incorrect assumptions about which TTS approaches are “best” for fast speech. We report on a cross-system comparison of the intelligibility of fast synthesized speech for users who are blind from birth. We used male and female voices from multiple TTS engines representing the main approaches to TTS: diphone synthesis, unit selection synthesis, HMM-based synthesis, and formant synthesis. We recruited participants from organizations that work with the blind. Each participant listened to and transcribed semantically underspecified sentences from a single TTS engine and voice, spoken at speeds ranging from 300 to 550 words/min. We used TTS engine, algorithm for speeding up synthesized speech, and speaker sex as independent variables, and transcription accuracy as the primary dependent variable. We report differences by TTS approach, speed up method, TTS engine type, and speaker sex.

Method and system for delivering text-to-speech in a real time telephony environment

A method and system for communicating information to a telephony user in response to a request for the information from the telephony user includes a text data source having text documents and a voice application operable with the text data source for retrieving a text document related to the requested information. A text-to-speech (TTS) manager divides the text document into text document segments. TTS engines convert the text document segments into speech segments. Each text document segment is converted into a speech segment by one TTS engine. A buffer reassembles the speech segments in the proper order and then delivers the speech segments to the user. The TTS manager divides the text document as a function of the rate at which speech segments are delivered to the user such that the speech segments are delivered from the TTS engines to the buffer and from the buffer to the user continuously.

Assessment of the intelligibility and perceived quality of speech produced by text-to-speech engines

The intelligibility and quality of speech produced by current text-to-speech (TTS) engines was assessed using well-established tests of speech intelligibility. TTS engines included Microsoft’s Whistler, AT&T’s Watson, and Elan Informatique’s ProVerbe. The speech intelligibility tests varied in complexity and in the degree to which syntax and lexical knowledge could aid perception. Stimuli included CVC words, semantically meaningful sentences, semantically anomalous sentences, and paragraphs representing different styles of discourse. Intelligibility was evaluated using transcription accuracy for isolated word and sentence items, as well as ratings of the clarity of articulation for word and paragraph items. Perceptual errors on word items also were analyzed by phoneme and word position (initial, medial, or final). Additional ratings were used to determine the perceived ‘‘naturalness’’ of meaningful sentences and paragraph items. Across levels of speech complexity, clear and consistent differences between TTS engines were obtained in terms of both intelligibility and naturalness. Possible explanations of the results will be discussed, as will potential areas where text-to-speech synthesis can generally be improved.