How to convert speech to text using Node.js

Translator|Li Rui

Reviewer|Sun Shujuan

This article will provide a brief overview of speech recognition technology and its common applications, and demonstrate a free speech-to-text API that Can be used to transcribe audio in MP3 and WAV file formats. This demo will include step-by-step instructions for calling this API using the "Run Node.js Code" example.

Overview of Speech Recognition Technology

People often think of speech recognition technology as a relatively new addition to the contemporary technology landscape, and that’s only part of the truth. . Speech recognition technology has been around for more than half a century, starting with limited-function number/text recognition systems developed by a few pioneer companies in the early 1950s. Although speech recognition technology has a long history and has been widely used in smart consumer devices over the past decade, speech recognition is still one of the more abstract technologies on the market today. This is because various speech recognition technologies span fields such as computer science, computational linguistics, and mathematics/statistics, and require large amounts of input from each field to obtain accurate speech-to-text results.

At a (very) high level, in order for speech recognition technology to perform its most basic tasks, a given audio file must first be preprocessed to optimize its quality. This is then broken down into smaller component signals and sorted. These sorted signals must be small enough so that a mathematical model can match them to certain phonemes (language-specific sounds that combine to create a word; such as "eeee" or "ahhh" noise), which helps match that Phrases or sentences in languages ​​are compared. Ultimately, the goal of a speech recognition service is simple: guess as accurately as possible what words were used in recorded audio, and continually improve and expand its language database until the guesses reach an acceptable level of accuracy.

This complex and inherently limited information guessing system makes even the most basic speech recognition services as dependent on language and dialect as they are on audio quality. Variations in language, accent, vocabulary, and the presence of background noise all create boundaries that are difficult for a single speech-to-text model to overcome. Furthermore, this complexity reflects the fundamental fact that speech recognition services are highly resource-intensive, relying on large and growing reference data sets for speech comparison, requiring considerable computing power to effectively utilize these data sets. Together, these factors make training a new speech-to-text model a daunting task.

Applications of Speech Recognition

Speech recognition is largely due to some ubiquitous innovations in the larger technology market, especially the near-infinite cloud computing data storage With the growth of solutions, speech recognition has become an efficient and useful consumer service that people now recognize in their daily lives. People can communicate and hold conversations directly using mobile phones, landlines and office devices to automatically query the information they need, record and organize their thoughts for later use, hear text messages read aloud, and more.

Currently, consumer applications are just the tip of the iceberg of speech recognition technology applications. Advances in the many interrelated processes of speech recognition create opportunities to expand audio transcription output and encourage an increasing number of commercial applications for speech-to-text conversion. There are many such examples around people. Many virtual meeting platforms now use speech recognition services to transcribe team presentations (often in real time), and the resulting text can be conveniently stored for anyone who might have missed the meeting. Chatbots use speech recognition to help people make choices on their phones, and the transcribed audio from these conversations can be used for something else: to inform better customer service practices in the future. Lectures, interviews, presentations and other speaking engagements that people may attend can be recorded and transcribed via personal devices to eliminate the labor of manually transcribing notes. For example, these transcriptions can be processed through natural language processing (NLP) models to find previously unseen (or unheard) insights from the transcribed text.

It’s worth mentioning that useful content moderation and SEO capabilities are also available from the scalable speech-to-text analytics service – this time in an enterprise data storage environment. Given the unreliability of content uploaded by most clients, automatically transcribing audio files uploaded to the website could easily adjust the language of the parent audio file to ensure it is safe for all listeners (e.g. to see if it contains highly controversial language, hate speech or any form of harassment of an individual or group). This significantly reduces the workload of content reviewers and increases their efficiency. These same transcription files can also be used to generate useful keywords, making audio files easier to search and retrieve from large databases.

Demonstration: CloudMersive Speech to Text API

One way to leverage speech recognition as a service is by including the Cloudmersive Speech to Text API. The API currently supports MP3 or WAV formats and uses a deep learning artificial intelligence model to provide high-precision audio transcription. Its API parameters are very simple, requiring only an audio file and a Cloudmersive API key (the API key can be obtained by signing up for a free account on its website, which generates up to 800 API calls per month). The following demonstrates how to build API calls using complementary Node.js code snippets.

The first step is to install the Node.js SDK. This can be done by running the following command:

<span style="color: rgb(89, 89, 89); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">npm</span> <span style="color: rgb(89, 89, 89); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">install</span> <span style="color: rgb(89, 89, 89); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">cloudmersive</span><span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">-</span><span style="color: rgb(89, 89, 89); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">speech</span><span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">-</span><span style="color: rgb(89, 89, 89); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">api</span><span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">-</span><span style="color: rgb(89, 89, 89); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">client</span> <span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">--</span><span style="color: rgb(89, 89, 89); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">save</span>

Alternatively, you can add the following snippet to its package.

<span style="color: rgb(102, 153, 0); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">"dependencies"</span>: {<br><span style="color: rgb(102, 153, 0); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">"cloudmersive-speech-client"</span>: <span style="color: rgb(102, 153, 0); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">"^1.1.5"</span><br>}

After the installation is complete, you can use the following code block to build API calls. At this time, you need to ensure that the following parameters are ready:

(1) MP3 or WAV audio file.

(2)CloudMersive API key.

<span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">var</span> <span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">CloudmersiveSpeechApiClient</span> <span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">=</span> <span style="color: rgb(89, 89, 89); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">require</span>(<span style="color: rgb(102, 153, 0); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">'cloudmersive-speech-api-client'</span>);<br><span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">var</span> <span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">defaultClient</span> <span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">=</span> <span style="color: rgb(89, 89, 89); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">CloudmersiveSpeechApiClient</span>.<span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">ApiClient</span>.<span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">instance</span>;<br><span style="color: rgb(106, 115, 125); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">// Configure API key authorization: Apikey</span><br><span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">var</span> <span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">Apikey</span> <span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">=</span> <span style="color: rgb(89, 89, 89); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">defaultClient</span>.<span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">authentications</span>[<span style="color: rgb(102, 153, 0); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">'Apikey'</span>];<br><span style="color: rgb(89, 89, 89); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">Apikey</span>.<span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">apiKey</span> <span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">=</span> <span style="color: rgb(102, 153, 0); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">'YOUR API KEY'</span>;<br><span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">var</span> <span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">apiInstance</span> <span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">=</span> <span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">new</span> <span style="color: rgb(89, 89, 89); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">CloudmersiveSpeechApiClient</span>.<span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">RecognizeApi</span>();<br><span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">var</span> <span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">speechFile</span> <span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">=</span> <span style="color: rgb(89, 89, 89); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">Buffer</span>.<span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">from</span>(<span style="color: rgb(89, 89, 89); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">fs</span>.<span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">readFileSync</span>(<span style="color: rgb(102, 153, 0); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">"C:\temp\inputfile"</span>).<span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">buffer</span>); <span style="color: rgb(106, 115, 125); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">// File | Speech file to perform the operation on.Common file formats such as WAV, MP3 are supported.</span><br><span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">var</span> <span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">callback</span> <span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">=</span> <span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">function</span>(<span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">error</span>, <span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">data</span>, <span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">response</span>) {<br><span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">if</span> (<span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">error</span>) {<br><span style="color: rgb(89, 89, 89); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">console</span>.<span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">error</span>(<span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">error</span>);<br>} <span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">else</span> {<br><span style="color: rgb(89, 89, 89); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">console</span>.<span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">log</span>(<span style="color: rgb(102, 153, 0); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">'API called successfully. Returned data: '</span> <span style="color: rgb(215, 58, 73); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">+</span> <span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">data</span>);<br>}<br>};<br><span style="color: rgb(89, 89, 89); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">apiInstance</span>.<span style="color: rgb(0, 92, 197); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">recognizeFile</span>(<span style="color: rgb(89, 89, 89); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">speechFile</span>, <span style="color: rgb(89, 89, 89); margin: 0px; padding: 0px; background: none 0% 0% / auto repeat scroll padding-box border-box rgba(0, 0, 0, 0);">callback</span>);

This completes the API construction without further code snippets. A successful API call will return a Textresult string containing the API transcription results. It is important to remember that the audio quality in the input file will have a significant impact on the API's ability to create accurate transcriptions, so it is recommended to pre-process and optimize the audio quality as much as possible before implementing this operation.

Original title: ​​How to Leverage Speech-to-Text With Node.js​​, author: Brian O'Neill

The above is the detailed content of How to convert speech to text using Node.js. For more information, please follow other related articles on the PHP Chinese website!