{"id":13988771,"url":"https://github.com/techytushar/voice2text","last_synced_at":"2025-06-20T07:37:54.480Z","repository":{"id":111152674,"uuid":"218344666","full_name":"techytushar/voice2text","owner":"techytushar","description":"Better and Smarter way to convert speech to text.","archived":false,"fork":false,"pushed_at":"2019-10-29T20:45:12.000Z","size":17,"stargazers_count":2,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-02-24T16:52:21.961Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"gpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/techytushar.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null}},"created_at":"2019-10-29T17:28:46.000Z","updated_at":"2021-11-14T19:09:46.000Z","dependencies_parsed_at":null,"dependency_job_id":"8f5dfcc9-7b55-43d5-b17c-9f3c6f043918","html_url":"https://github.com/techytushar/voice2text","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/techytushar/voice2text","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/techytushar%2Fvoice2text","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/techytushar%2Fvoice2text/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/techytushar%2Fvoice2text/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/techytushar%2Fvoice2text/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/techytushar","download_url":"https://codeload.github.com/techytushar/voice2text/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/techytushar%2Fvoice2text/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":260901852,"owners_count":23079849,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-08-09T13:01:21.114Z","updated_at":"2025-06-20T07:37:49.471Z","avatar_url":"https://github.com/techytushar.png","language":"Python","funding_links":[],"categories":["Python"],"sub_categories":[],"readme":"# Voice2Text\n\n## Task 1\n\nThe task was to create a reusable library that can convert speech to cleaned text.\n\n### How to use\n\nThe module works with Python \u003e= 3.6. It is made in the form of a Python package. Inside the cloned folder run the following command to install the package:\n\n```shell\npip install .\n```\n\nNow open the Python shell and use as below:\n\n```python\nimport voice2text as v\nv.convert()\n```\n\nThe `convert()` function will listen to your message and return you the cleaned text. Make sure you have a working internet connection.\n\n\n### Approach\nI used the Python's `speech_recognition` module to convert the speech to text and then processed the text to convert it.  \nTo solve the task I created 4 functions\n\n* First function does the basic text cleaning.\n* Then the currency words are converted to symbols, such as `10 dollars` is converted to `$10`. It converts the word into its symbol only when the word is found with a number. For ex. it will convert `I have 10 dollars` to `I have $10` but the sentence `Euro is stronger than dollar` will remain unchanged.\n* Then the title words are converted to their acronyms such as `doctor` to `Dr.`. In this also if a title appears then it checks if it appears with a name of a person (using Named Entity Recognition) then only it converts it into its acronym.\n* The last one converts the repetitions. Such as `double A` is converted to `AA`\n\n## Task 2\n\nTo design a spoken English to written English conversion system that can be continuously matured overtime.\n\n### Approach\n\nConverting speech to text is a trivial task and can be done using the existing APIs with good accuracy. The main challenge lies in the conversation, to get a text from the user and generating a valid response. This is challenging because there are numerous rules and exceptions present in every language and all those are very hard to capture.\n\nFor building a conversational bot, the best approach would be to construct a deep learning RNN model and train it on a dataset. After training we can evaluate the performance and fine tune it. Using the various NLP libraries we can hard-code some of the conversational rules and clean the text input by the user. Then feed the cleaned text to the bot to generate better response.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftechytushar%2Fvoice2text","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ftechytushar%2Fvoice2text","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ftechytushar%2Fvoice2text/lists"}