Icon

01_​Example_​Speech-to-Text

There has been no title set for this workflow's metadata.

KNIME Audio Nodes - And Example of Speech-to-Text

This workflow shows the functionality of the KNIME Audio nodes in combination with Text Mining.

The example audio file contains a one minute sample of an audio book of Alice's Adventures in Wonderland by Lewis Carroll.

- Read in audio files
- Use speech-to-text services
- Use text mining to see which speech-to-text service delivers the best results

NOTE:
- The CMUSphinx4 speech recognition service does not require any credentials
- Bing and IBM Watson require access to the respective services and the corresponding credentials

URL: Text Processing with KNIME https://www.knime.com/knime-text-processing
URL: Audio file source https://etc.usf.edu/lit2go/1/alices-adventures-in-wonderland/1/chapter-i-down-the-rabbit-hole/

This workflow shows the functionality of the KNIME Audio nodes in combination with Text Mining. It tests different text-to-speech services are tested and compares the results. Read audio data, convert speech to text Text Mining - Preprocessing Text Mining - Transformation and Similarity Search Open speech-to-text service,no credentials requiredRead audio dataNeeds credentialsNeeds credentialsAdd source identifierAdd source identifierAdd source identifierConvert to documentcellOriginal transcriptof audio fileAdd rowwith originaltranscriptKeep onlydocument columnTransformto Bag of WordsCase converter,punctuationeraserRelative termfrequencyBinary vectorcreationCompare speechrecognition results tooriginal transcriptOriginalTranscriptRemove sourcesfrom documenttextExtract sourceidentifierFilter documentvector columnsDownload theaudio fileBlend results fromeach serviceNode 63 CMUSphinx4 SR List Audio Files Bing SR IBM Watson SR ConstantValue Column ConstantValue Column ConstantValue Column Column Rename(deprecated) Column Rename(deprecated) Column Rename(deprecated) Strings To Document(deprecated) Table Creator Column Filter Concatenate Column Filter Bag of Words Creator(deprecated) Preprocessing TF Document vector(deprecated) Distance MatrixCalculate Similarity Search Row Filter Dictionary Filter Category to Class Column Filter Explorer Browser Concatenate Column Filter Excel Writer This workflow shows the functionality of the KNIME Audio nodes in combination with Text Mining. It tests different text-to-speech services are tested and compares the results. Read audio data, convert speech to text Text Mining - Preprocessing Text Mining - Transformation and Similarity Search Open speech-to-text service,no credentials requiredRead audio dataNeeds credentialsNeeds credentialsAdd source identifierAdd source identifierAdd source identifierConvert to documentcellOriginal transcriptof audio fileAdd rowwith originaltranscriptKeep onlydocument columnTransformto Bag of WordsCase converter,punctuationeraserRelative termfrequencyBinary vectorcreationCompare speechrecognition results tooriginal transcriptOriginalTranscriptRemove sourcesfrom documenttextExtract sourceidentifierFilter documentvector columnsDownload theaudio fileBlend results fromeach serviceNode 63CMUSphinx4 SR List Audio Files Bing SR IBM Watson SR ConstantValue Column ConstantValue Column ConstantValue Column Column Rename(deprecated) Column Rename(deprecated) Column Rename(deprecated) Strings To Document(deprecated) Table Creator Column Filter Concatenate Column Filter Bag of Words Creator(deprecated) Preprocessing TF Document vector(deprecated) Distance MatrixCalculate Similarity Search Row Filter Dictionary Filter Category to Class Column Filter Explorer Browser Concatenate Column Filter Excel Writer

Nodes

Extensions

Links

Download

To use this workflow in KNIME, download it from the below URL and open it in KNIME:

Download Workflow
Created by: rb
Created at: 2017-06-28
On NodePit since: 2023-12-13
Last update: 2024-04-27
Created with KNIME version: v5.2.0
Tags: text miningcloudCMUSphinx4 SRBing SRIBM Watson SRspeech-to-text

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!