В работе

Embedded Offline ASR Engine Development For English and Chinese Languages

This project is to use Python and Mozilla’s DeepSpeech ASR (automatic speech recognition) engine on different platforms (such as Raspberry Pi 4 - 1 GB, Nvidia Jetson Nano, Windows PC, and Linux PC, Samsung Galaxy A50, Huawei P20) in order to develop a refined ASR engine for English and Chinese (Mandarin) languages with following functionalities, with source code, instruction, data and an API documentation delivered after development complete. The architecture of Deep Speech is an end-to-end trainable, character-level, deep recurrent neural network (RNN). It is a deep neural network with recurrent layers that gets audio features as input and outputs characters directly — the transcription of the audio, and uses LSTM (Long short-term memory) cells instead of GRU (gated recurrent unit) cells. This project targets <6% of Word Error Rate, and especially for key phrases and key words to <3% WER, close to human level performance for English and Chinese languages.

1. Use the latest Mozilla’s DeepSpeech ASR engine which comes with .tflite model (TensorFlow Lite), faster than real-time on a single core of a Raspberry Pi 4, and able to make our own audio transcription application with hot word detection function.

2. Generate Confidence Scores of high accuracy level. Be able to retrieve result confidence scores for each English word and for each English sentence in a transcription of audio into text (confidence score for both word and sentence level), and they should be highly reliable.

3. Support Keyword Spotting modes that can recognize in a continuous stream. User can configure a list of key phrases to search for and specify the detection threshold for each of them. This mode should reliably work in continuous speech stream and can be used for keyword activation. Equivalent to pocketsphinx -kws and –keyphrase options (The methods are ps_set_keyphrase and ps_set_kws).

4. Recognize accurately at least a thousand of commands and controls that user can define in a simple text editor in keyword spotting mode or keyword activation mode. (same as No. 3 requirement, with 3% WER)

5. Mozilla’s DeepSpeech ASR word error rate on LibriSpeech’s test-clean set is 6.5%, which we target to improve to 6% by this project, and for key phrases and key words to <3% WER, close to human level performance.

6. Use automatic phoneme alignment, VAD and other methods to detect the start time, the end time and position for each recognized phoneme, word and sentence, with an output data structure readable at real-time for the complete set of phoneme position information. Forced alignment refers to the process by which orthographic transcriptions are aligned to audio recordings to automatically generate phone level segmentation. As described at the following link:[login to view URL]。

When developing an ASR system, “good initial estimates … are essential” when training Gaussian Mixture Model (GMM) parameters (Rabiner and Juang, 1993, p. 370). Phoneme location information is also critical when building concatenative text-to-speech systems.

7. Implement CTC decoder as an important optimization: integrating appropriate language model into the decoder.

Навыки: Python, Tensorflow

Показать больше: translation fee english chinese, easy translations english chinese, translate english chinese business websites, english chinese free, english chinese generator, 100 convert english chinese translation, translation cooperation english chinese translation, english chinese wanted, accurate english chinese translator, english chinese frelance translation, english chinese languages translation marketing, looking for a chinese translater for english literature, looking for a assistant job in turkey for english chinese and turkish speaking man, looking for freelance english chinese translation job, english chinese translator for hire

О работодателе:
( 3 отзыв(-а, -ов) ) Flushing, United States

ID проекта: #24813208



Hello! I'd like to deliver asr engine for mobile platform. I'm familiar with CTC loss, Speech recognition, DeepSpeech, pytorch, tflite. I'll do the job blazingly fast. Please, give me a try!

$1792 USD за 45 дней(-я)
(6 отзывов(-а))

7 фрилансеров(-а) готовы выполнить эту работу в среднем за $1335


Hi there, I have read the brief details on the job listing. You can check my experience, customer feed backs and my portfolio here: https://www.freelancer.com/u/AwaisChaudhry?w=f I believe its a doable job I have grea Больше

$750 USD за 28 дней(-я)
(5 отзывов(-а))

Alert: I will give you 20% discount on my bid rate also give on my All Services. So grabs this special offer is limited. Let’s get to the point. I am an highly experienced freelancer . I am offering services in Ardu Больше

$250 USD за 7 дней(-я)
(11 отзывов(-а))

Hello. My name is Mohsen and thanks for posting the job. I have read all your requirements for 'Embedded Offline ASR Engine Development For English and Chinese Languages' and I fully understood it. I've already done th Больше

$555 USD за 3 дней(-я)
(2 отзывов(-а))

Hello! I would love to help you reach your goals on-time and on-budget. I have extensive experience creating web platforms from simple informational sites, to high-performance Single Page Applications Больше

$748 USD за 7 дней(-я)
(3 отзывов(-а))

Hi, dear Bei I have read your description carefully. Very interesting job but it may take long time, and high cost. Can we contact and discuss over chat? If you can proceed this project in hourly mode, I can work for Больше

$5000 USD за 30 дней(-я)
(5 отзывов(-а))

Hi, I am an Electrical, Electronics and Embedded Engineer. A PCB Designer, Arduino/Raspberry Pi, ESP32, ESP8266 and internet of things expert. I read through the job description very carefully and I am absolutely sure Больше

$250 USD за 7 дней(-я)
(0 отзывов(-а))