Hi，
\n\nAs a newbie in SR, I collected some references these days. Maybe help for you.
\n\nHi，
\n\nAs a newbie in SR, I collected some references these days. Maybe help for you.
\n\nAwesome! Thanks for collecting these materials!
\n\nI would like to add a little bit:
\n\nthank you. these links are very helpful.
", "rawMarkdown": "thank you. these links are very helpful.", "isDeleted": true } ] }, { "id": 250575, "postDate": "2017-11-30T05:35:29.700Z", "content": "If you're interested in learning more on the DSP side, here are some links I've found useful:
\n\nHere are a few other useful links:
\n\nHope these are useful!
", "rawMarkdown": "If you're interested in learning more on the DSP side, here are some links I've found useful:\n\n* Archived MIT edx course taught by Alan Oppenheim, a legend in DSP: https://www.edx.org/course/discrete-time-signal-processing-mitx-6-341x-1 \n* Coursera course on analyzing music has a lot of overlap with analyzing speech: https://www.coursera.org/learn/audio-signal-processing : \n- Audio Content Analysis book : https://www.audiocontentanalysis.org . This is on O'Reilly Safari if you have a subscription (or through ACM).\n\nHere are a few other useful links:\n\n* Video from Andrew Coates on Deep speech, really well explained and useful for this competition.: https://youtu.be/g-sndkf7mCs \n* Github repo for above: https://github.com/baidu-research/ba-dls-deepspeech . It has Tensorflow and Theano code including the CTC algorithm.\n* Mozilla also have an open source voice dataset. They implemented deep speech in this github repo: https://github.com/mozilla/DeepSpeech\n\nHope these are useful!\n", "votes": 2 }, { "id": 249563, "postDate": "2017-11-28T18:40:16.480Z", "content": "Great resources, thanks!
\n\nHere's another paper I found useful: Very deep convolutional neural networks for robust speech recognition
\n\nIn the paper, the authors discussed different choices of CNN architecture and compared their results obtained by each option.
", "rawMarkdown": "Great resources, thanks!\n\nHere's another paper I found useful: [Very deep convolutional neural networks for robust speech recognition](https://arxiv.org/pdf/1610.00277)\n\nIn the paper, the authors discussed different choices of CNN architecture and compared their results obtained by each option.", "votes": 2, "replies": [ { "id": 262579, "postDate": "2017-12-27T06:13:22.647Z", "content": "This is very useful, thanx a lot!\nBut still I could not catch why the input size in the paper is 11 or 17 at time axis. It seems to be a very short frame!
", "rawMarkdown": "This is very useful, thanx a lot!\nBut still I could not catch why the input size in the paper is 11 or 17 at time axis. It seems to be a very short frame!\n" } ] }, { "id": 248109, "postDate": "2017-11-24T20:40:21.210Z", "content": "Good references, thx
", "rawMarkdown": "Good references, thx" }, { "id": 251000, "postDate": "2017-11-30T15:25:54.947Z", "rawMarkdown": "", "isDeleted": true }, { "id": 248534, "postDate": "2017-11-26T11:50:07.583Z", "rawMarkdown": "", "isDeleted": true, "replies": [ { "id": 248584, "postDate": "2017-11-26T14:45:06.227Z", "content": "\" Transforms a spectrogram into a form that's useful for speech recognition.\"
\n\n", "rawMarkdown": "\" Transforms a spectrogram into a form that's useful for speech recognition.\"\n\nhttps://github.com/tensorflow/tensorflow/blob/ab0fcaceda001825654424bf18e8a8e0f8d39df2/tensorflow/examples/speech_commands/input_data.py#L398" } ] }, { "id": 1044955, "postDate": "2020-10-10T08:43:47.093Z", "content": "Thanks for sharing.
", "rawMarkdown": "Thanks for sharing.\n\n" }, { "id": 1044781, "postDate": "2020-10-10T05:35:01.203Z", "content": "Thanks for sharing.
", "rawMarkdown": "Thanks for sharing. " }, { "id": 484589, "postDate": "2019-03-06T07:57:24Z", "content": "Very helpful links. Thank you!
", "rawMarkdown": "Very helpful links. Thank you!" }, { "id": 252445, "postDate": "2017-12-02T23:42:46.393Z", "content": "Thanks for sharing, whiteworld
", "rawMarkdown": "Thanks for sharing, whiteworld" }, { "id": 249243, "postDate": "2017-11-28T02:02:24.607Z", "content": "Thanks for sharing...
", "rawMarkdown": "Thanks for sharing..." }, { "id": 248355, "postDate": "2017-11-25T20:06:30.197Z", "content": "Thanks for sharing!
", "rawMarkdown": "Thanks for sharing!" }, { "id": 248102, "postDate": "2017-11-24T19:59:52.107Z", "content": "Great share! Thanks for the post!
", "rawMarkdown": "Great share! Thanks for the post!" } ], "comments": [ { "id": 247687, "author_name": "Shujian Liu", "author_url": "", "post_date": "2017-11-23T16:07:26.220000", "content": "Awesome! Thanks for collecting these materials!
\n\nI would like to add a little bit:
\n\nthank you. these links are very helpful.
", "votes": 0, "replies": [] } ] }, { "id": 250575, "author_name": "tb303", "author_url": "", "post_date": "2017-11-30T05:35:29.700000", "content": "If you're interested in learning more on the DSP side, here are some links I've found useful:
\n\nHere are a few other useful links:
\n\nHope these are useful!
", "votes": 2, "replies": [] }, { "id": 249563, "author_name": "Zhixing Wang", "author_url": "", "post_date": "2017-11-28T18:40:16.480000", "content": "Great resources, thanks!
\n\nHere's another paper I found useful: Very deep convolutional neural networks for robust speech recognition
\n\nIn the paper, the authors discussed different choices of CNN architecture and compared their results obtained by each option.
", "votes": 2, "replies": [ { "id": 262579, "author_name": "Ivan Timoshilov", "author_url": "", "post_date": "2017-12-27T06:13:22.647000", "content": "This is very useful, thanx a lot!\nBut still I could not catch why the input size in the paper is 11 or 17 at time axis. It seems to be a very short frame!
", "votes": 0, "replies": [] } ] }, { "id": 248109, "author_name": "wolfgang", "author_url": "", "post_date": "2017-11-24T20:40:21.210000", "content": "Good references, thx
", "votes": 0, "replies": [] }, { "id": 251000, "author_name": "", "author_url": "", "post_date": "2017-11-30T15:25:54.947000", "content": "", "votes": 0, "replies": [] }, { "id": 248534, "author_name": "", "author_url": "", "post_date": "2017-11-26T11:50:07.583000", "content": "", "votes": 0, "replies": [ { "id": 248584, "author_name": "whiteworld", "author_url": "", "post_date": "2017-11-26T14:45:06.227000", "content": "\" Transforms a spectrogram into a form that's useful for speech recognition.\"
\n\n", "votes": 0, "replies": [] } ] }, { "id": 1044955, "author_name": "Bala", "author_url": "", "post_date": "2020-10-10T08:43:47.093000", "content": "Thanks for sharing.
", "votes": 0, "replies": [] }, { "id": 1044781, "author_name": "AjayJangid", "author_url": "", "post_date": "2020-10-10T05:35:01.203000", "content": "Thanks for sharing.
", "votes": 0, "replies": [] }, { "id": 484589, "author_name": "xtian", "author_url": "", "post_date": "2019-03-06T07:57:24", "content": "Very helpful links. Thank you!
", "votes": 0, "replies": [] }, { "id": 252445, "author_name": "Ravi Teja Gutta", "author_url": "", "post_date": "2017-12-02T23:42:46.393000", "content": "Thanks for sharing, whiteworld
", "votes": 0, "replies": [] }, { "id": 249243, "author_name": "Sam Zhou", "author_url": "", "post_date": "2017-11-28T02:02:24.607000", "content": "Thanks for sharing...
", "votes": 0, "replies": [] }, { "id": 248355, "author_name": "Renzo", "author_url": "", "post_date": "2017-11-25T20:06:30.197000", "content": "Thanks for sharing!
", "votes": 0, "replies": [] }, { "id": 248102, "author_name": "abkosar", "author_url": "", "post_date": "2017-11-24T19:59:52.107000", "content": "Great share! Thanks for the post!
", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "247611": "Hi，\n\nAs a newbie in SR, I collected some references these days. Maybe help for you.\n\n### Articles\n- [https://www.tensorflow.org/versions/master/tutorials/audio_recognition][1]\n- [https://www.kaggle.com/davids1992/speech-visualization-and-investigation][2]\n- [How to do Speech Recognition with Deep Learning][3]\n- [A quick hack to align single-word audio recordings][4]\n- [Speech Processing for Machine Learning: Filter banks, Mel-Frequency Cepstral Coefficients (MFCCs) and What's In-Between][5]\n- [Speech Recognition: You down with CTC?][6]\n\n### Papers\n- [ Convolutional Neural Networks for Small-footprint Keyword Spotting][7]\n- [small-footprint keyword spotting using deep neural networks][8]\n- [Deep Speech: Scaling up end-to-end speech recognition][9]\n- [Connectionist Temporal Classification: Labelling Unsegmented Sequence Data with Recurrent Neural Networks][10]\n- [Towards End-to-End Speech Recognition with Recurrent Neural Networks][11]\n\n### Videos\n- [Deep Learning for Speech Recognition (Adam Coates, Baidu)][12]\n- [How to Make a Simple Tensorflow Speech Recognizer][13]\n\n\n [1]: https://www.tensorflow.org/versions/master/tutorials/audio_recognition\n [2]: https://www.kaggle.com/davids1992/speech-visualization-and-investigation\n [3]: https://medium.com/@ageitgey/machine-learning-is-fun-part-6-how-to-do-speech-recognition-with-deep-learning-28293c162f7a\n [4]: https://petewarden.com/2017/07/17/a-quick-hack-to-align-single-word-audio-recordings/\n [5]: http://haythamfayek.com/2016/04/21/speech-processing-for-machine-learning.html\n [6]: https://gab41.lab41.org/speech-recognition-you-down-with-ctc-8d3b558943f0\n [7]: http://www.isca-speech.org/archive/interspeech_2015/papers/i15_1478.pdf\n [8]: https://static.googleusercontent.com/media/research.google.com/zh-CN//pubs/archive/42537.pdf\n [9]: https://arxiv.org/pdf/1412.5567.pdf\n [10]: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.75.6306&rep=rep1&type=pdf\n [11]: http://proceedings.mlr.press/v32/graves14.pdf\n [12]: https://www.youtube.com/watch?v=g-sndkf7mCs\n [13]: https://www.youtube.com/watch?v=u9FPqkuoEJ8", "247687": "Awesome! Thanks for collecting these materials!\n\nI would like to add a little bit:\n\n - Awesome-tensorflow: https://github.com/jtoy/awesome-tensorflow\n - Awesome-rnn: https://github.com/kjw0612/awesome-rnn\n - Awesome-speech-recognition-speech-synthesis-papers: https://github.com/zzw922cn/awesome-speech-recognition-speech-synthesis-papers\n\n\n", "250575": "If you're interested in learning more on the DSP side, here are some links I've found useful:\n\n* Archived MIT edx course taught by Alan Oppenheim, a legend in DSP: https://www.edx.org/course/discrete-time-signal-processing-mitx-6-341x-1 \n* Coursera course on analyzing music has a lot of overlap with analyzing speech: https://www.coursera.org/learn/audio-signal-processing : \n- Audio Content Analysis book : https://www.audiocontentanalysis.org . This is on O'Reilly Safari if you have a subscription (or through ACM).\n\nHere are a few other useful links:\n\n* Video from Andrew Coates on Deep speech, really well explained and useful for this competition.: https://youtu.be/g-sndkf7mCs \n* Github repo for above: https://github.com/baidu-research/ba-dls-deepspeech . It has Tensorflow and Theano code including the CTC algorithm.\n* Mozilla also have an open source voice dataset. They implemented deep speech in this github repo: https://github.com/mozilla/DeepSpeech\n\nHope these are useful!\n", "249563": "Great resources, thanks!\n\nHere's another paper I found useful: [Very deep convolutional neural networks for robust speech recognition](https://arxiv.org/pdf/1610.00277)\n\nIn the paper, the authors discussed different choices of CNN architecture and compared their results obtained by each option.", "248109": "Good references, thx", "251000": "", "248534": "", "1044955": "Thanks for sharing.\n\n", "1044781": "Thanks for sharing. ", "484589": "Very helpful links. Thank you!", "252445": "Thanks for sharing, whiteworld", "249243": "Thanks for sharing...", "248355": "Thanks for sharing!", "248102": "Great share! Thanks for the post!" } }