Eines per a l’anàlisi acústica i la transcripció de la parla

Eines per a l’anàlisi acústica de la parla

Eines per a la transcripció i l’etiquetatge de corpus orals

Anvil, M. Kipp, University of Applied Sciences, Augsburg


«ANVIL is a free video annotation tool, developed by Michael Kipp. It offers multi-layered annotation based on a user-defined coding scheme. During coding the user can see color-coded elements on multiple tracks in time-alignment. Some special features are cross-level links, non-temporal objects, timepoint tracks, coding agreement analysis, 3D viewing of motion capture data and a project tool for managing whole corpora of annotation files. Originally developed for gesture research in 2000, ANVIL is now being used in many research areas including human-computer interaction, linguistics, ethology, anthropology, psychotherapy, embodied agents, computer animation and oceanography.

ANVIL can import data from phonetic tools like PRAAT which allow precise and comfortable speech transcription (see my video tutorial). Anvil can display waveform and pitch contour. Anvil's data files are XML-based. Exported tables can be used for analysis in statistical toolkits like SPSS or Statistica. The coming version will also be able to import ELAN files. ANVIL is written in Java and runs on Windows, Macintosh and Unix platforms.»



Audiamus, N. Thieberger, School of Language and Linguistics, University of Melbourne


«A tool for building corpora of linked transcripts and digitised media.

Audiamus instantiates the links to digitised media. It requires no segmentation of the sound/video file. Currently there is no limit to the size of the media file or the number of transcripts. Each ’card’ of the current model represents a single transcript (typically a complete side of a cassette). Time-aligned transcripts, as produced for example by SoundIndex or Transcriber are the input for Audiamus.

The transcripts in Audiamus are plain text and can be edited, as can the timecodes. Thus the data in Audiamus is the master copy of the transcript that is improved incrementally with use. To avoid the problem of data being locked up in proprietary formats there is a mass export function that dumps all linked text and timecodes to plain text files, or to whatever format the user selects.»


CSL, Computerized Speech Lab, Pentax Medical


«CSL is the most comprehensive PC-based system available for speech acquisition, analysis, editing, and playback. An integrated hardware/software system, the versatile platform is recognized internationally by both clinicians and researchers for its unique combination of sophistication, flexibility, and ease-of-use.

The system’s robust hardware meets the rigorous specifications required by speech professionals and researchers. It contains an external module for high-fidelity data acquisition (>86 dB dynamic range), DSP circuitry for real-time processing/display of speech parameters needed for therapy applications, and CD-quality playback for critical listening tasks. The core software is fully integrated with the hardware. It contains a rich set of easily applied analysis and editing features and is complemented by 15 application specific (e.g., clinical, linguistic, etc.) software modules and databases.

Built on Kay’s decades of experience in speech analysis, the CSL accommodates the many and varied needs of speech/voice clinicians, phoneticians, speech scientists, phoniatricians, and otolaryngologists. CSL was developed jointly with Speech Technology Research (STR) of Victoria, B.C., Canada

Current CSL options include:


CSLU Toolkit, Center for Spoken Language Understanding, Oregon Graduate Institute

«The CSLU Toolkit has been supporting research, development and learning activities for spoken language systems since January, 1996. It is designed to support a wide range of research activities, including data capture and analysis, corpus development, research in multilingual recognition and understanding, dialogue design, speech synthesis speaker recognition and language recognition, among others. In addition, the Toolkit provides easy to use graphical authoring tools (CSLUrp) for rapid prototyping of spoken language systems for useful applications. Finally, the toolkit is designed to provide a good environment for learning about spoken language technology. The Toolkit has been used to teach short courses, and students taking these courses have produced novel and useful spoken language systems, as described on our short course page.

The Toolkit currently runs on Unix platforms which have Tcl/Tk (freely available).»


Dolmen, J. Eychenne, Hankuk University of Foreign Studies


«Note: as of November 2019, Dolmen is superseded by Phonometrica.

Dolmen is a free, open-source software toolbox for data analysis in linguistics. It offers a user-friendly interface to manage, annotate and query language corpora. It is particularly well suited for dealing with time-aligned data. The main features it offers are:

Dolmen runs on all major platforms (Windows, Mac OS X and GNU/Linux) and is freely available under the terms of the GNU General Public License (GPL).»


ELAN, EUDICO Linguistic Annotator, Max Plank Institute for Psycholinguistics


«With ELAN a user can add an unlimited number of textual annotations to audio and/or video recordings. An annotation can be a sentence, word or gloss, a comment, translation or a description of any feature observed in the media. Annotations can be created on multiple layers, called tiers. Tiers can be hierarchically interconnected. An annotation can either be time-aligned to the media or it can refer to other existing annotations. The content of annotations consists of Unicode text and annotation documents are stored in an XML format (EAF).

Main features and characteristics of ELAN:


Max Planck Institute for Psycholinguistics. (2019). ELAN (Versió 5.8) [Programa informàtic]. The Language Archive. https://archive.mpi.nl/tla/elan


Max Planck Institute for Psycholinguistics. (2019). ELAN (Versió 5.8) [Programa informàtic]. The Language Archive. https://archive.mpi.nl/tla/elan


GIPOS, Institute for Perception Research, Eindhoven


«GIPOS stands for Graphical Interactive Processing of Speech. It is an integrated speech processing program. It provides the tools you need to create, view, play and manipulate waveforms, spectrograms and other forms of speech data. You’ll find:

GIPOS runs on SiliconGraphics and SUN workstations under Unix and on PCs under MS-Windows95 and Linux.»


ISA, Intelligent Speech Analyser, Oy Pitchsystems

«The main scopes of application include:

All the analysis programs have been written using a machine language, because in this way ISA is many times faster than using a high level language. ISA is the unique software in the world. The use of ISA is very simple. All the analyses have their own windows. All the functions of the ISA are controlled by the mouse. All the displays can be listened to. ISA-software is running in Apple Macintosh computer.»


LaBB-CAT, R. Fromont i J. Hay, New Zealand Institute of Language, Brain and Behaviour


«LaBB-CAT is a browser-based linguistics research tool that stores audio or video recordings, text transcripts, and other annotations.

Annotations of various types can be automatically generated or manually added.

The transcripts and annotations can be searched for particular text or regular expressions. The search results, or entire transcripts, can be viewed or saved in a variety of formats, and the related parts of the recordings can be played or opened in acoustic analysis software, all directly through the web-browser.



lingWaves, WEVOSYS


«lingWAVES has become one of the most used system for professional voice and speech analysis, biofeedback and documentation in the last years. A combination of standard and new technology analysis and processing together with an easy handling are the key features of this unique system.

lingWAVES module puzzle: The system consists of different modules managed by the lingWAVES basis user interface. A client manager allows a patient/client based analysis and documentation with the benefit of comparing and tracking results over time.

The modular character of lingWAVES allows to offer different module combinations (suites) so that a wide range of professional users can use the system, starting from speech and language therapy, over Otolaryngology /ENT up to services for professional singers and speakers. You can also upgrade a lingWAVES module at any time.

System Requirements: Windows 10 ( Mac OS with Boot Camp and installed Windows OS 10).»


MATLAB Signal Processing Toolbox, The Math Works


«The Signal Processing Toolbox provides a rich, customizable framework for digital signal processing (DSP). Built on a solid foundation of filter design and spectral analysis techniques, the toolbox contains powerful tools for algorithm development, signal and linear system analysis, and time-series data modeling. The toolbox is useful in applications such as speech and audio processing, communications, geophysics, real-time control, finance, radar, and medicine.

Signal and linear system models:


MelAn, J. M. Garrido, Departamento de Lengua Española y Lingüística General, Universidad Nacional de Educación a Distancia


«MelAn is a tool for the automatic stylisation, annotation and modelling of F0 contours. It is made of a set of Praat and R scripts that perform the tasks of F0 stylisation, labelling and modelling. They can be run on Windows, Mac OS X or Linux. It has been conceived for the automatic processing and analysis of large corpora.

The tool applies automatically the framework and methodology for the analysis of F0 contours proposed in Garrido (1996, 2001). This procedure is intended to obtain a symbolic representation of F0 contours which captures their perceptually relevant features, in the sense that it should be possible to build a ‘synthetic’ contour from the symbolic representation almost identical to the original contour from a perceptual point of view.

MelAn is available for public download.»


PCquirerX, Scicon R&D


«PCquirer & Macquirer features include:



Phon, Y. Rose i G. Hedlund, The PhonBank Project, CHILDES (Child Language Data Exchange System), TalkBank


«Phon is a software program that greatly facilitates a number of tasks related to the analysis of transcript-based and acoustically-measured speech data. Built to support research in phonological development (including babbling), second language acquisition, and phonological disorders, Phon can also be used for virtually all types of phonological investigations (e.g. loanword phonology, fieldwork in phonology, sociolinguistic studies). Phon supports multimedia data linkage, unit segmentation (e.g. utterance, word), multiple-blind transcription, automatic labeling of data (features, syllabification), and systematic comparisons between target (model) and actual (produced) phonological forms. Phon is also equipped with many facilities for data analysis, including query methods for phonology (e.g. phones, features, syllables, ...) as well as acoustic data.

Versió 2 of Phon brings together two of the most important areas of empirical investigation in the area of child phonology, as it integrates transcript-based analyses of phonological data with the facilities for acoustic analysis provided by Praat. With this new version of Phon, and in addition to the functions listed above, the user can now:

All of these functions are accessible through a user-friendly graphical interface. Databases managed within Phon can also be queried using a powerful search system adapted for the needs of the phonologist. This software program works on Mac OS X, Windows and Linux platforms and is compliant with the CHILDES (TalkBank) XML data format. Phon is being made freely available to the community as open-source software. Phon facilitates data exchange among researchers and is currently used for the elaboration of the shared PhonBank database, designed to support empirical needs of research in all areas of phonology and phonological development.»


Rose, Y. i MacWhinney, B. (2014). The PhonBank project: Data and software-assisted methods for the study of phonology and phonological development. A J. Durand, U. Gut i G. Kristoffersen (Ed.), The Oxford handbook of corpus phonology (p. 308–401). Oxford University Press. https://doi.org/10.1093/oxfordhb/9780199571932.013.023


Rose, Y. i MacWhinney, B. (2014). The PhonBank project: Data and software-assisted methods for the study of phonology and phonological development. A J. Durand, U. Gut i G. Kristoffersen (Ed.), The Oxford handbook of corpus phonology (p. 308–401). Oxford University Press. https://doi.org/10.1093/oxfordhb/9780199571932.013.023


PHONEDIT Signaix, Laboratoire Parole et Langage, Université de Provence


«PHONÉDIT is a signal editor that permits to record, edit, labelize, and analyse various types of signals. This software is dedicated to speech analysis. However it has the capability to analyse also aerodynamic parameters, electro-palatographic frames and kynesiographic movements.

It reads and writes the most common kind of file formats like MS-WAVE, CSL, Signalyze, ASCII, or raw binaries.

Many functions are applicable on the edited signals:

It runs with recent Windows operating system on PC.

PHONEDIT Signaix is free of charge and can be downloaded.»


Phonometrica, J. Eychenne (Hankuk University of Foreign Studies) i L. Courdès-Murphy (Universität Wien).


«Phonometrica is a free, open-source software platform for the annotation and analysis of speech corpora. It offers a user-friendly interface to manage, annotate, analyze and query language corpora. It is particularly well suited for dealing with time-aligned data. The main features it offers are:

Phonometrica runs on all major platforms (Windows, macOS and GNU/Linux) and is freely available under the terms of the GNU General Public License.»


Eychenne, J. i Courdès-Murphy, L. (2019). Phonometrica (Versió 0.7.5) [Programa informàtic]. http://www.phonometrica-ling.org


PitchWorks, Scicon R&D



Praat, P. Boersma i D. Weenink, Institute of Phonetic Sciences, University of Amsterdam


«Speech analysis:

Speech synthesis:

Listening experiments:

Labelling and segmentation:

Speech manipulation:

Learning algorithms:






Macintosh, Windows, Linux, Chromebook, Raspberry Pi (FreeBSD, SGI, Solaris, HPUX)»


Boersma, P. i Weenink, D. (2021). Praat: Doing phonetics by computer (Versió 6.1.40) [Programa informàtic]. University of Amsterdam. http://www.praat.org

Scripts per a Praat

Department of Linguistics. (2020). Software and resources. University of Victoria. https://www.uvic.ca/humanities/linguistics/resources/software/index.php

Dellwo, V. (2019, 20 de novembre). Material. Universität Zürich, Institut für Computerlinguistik. https://www.cl.uzh.ch/de/people/team/phonetics/vdellw/software.html

DiCanio, C. (s. d.). Scripts. University at Buffalo, Department of Linguistics. https://www.acsu.buffalo.edu/~cdicanio/scripts.html

Felty, R. (s. d.). Praat scripts. robfelty.com. https://robfelty.com/academic/praat-scripts

Kawahara, S. (s. d.). Scripts. http://user.keio.ac.jp/~kawahara/resource.html

Lennes, M. (2017, 8 de març). SpeCT – Speech Corpus Toolkit for Praat (v1.0.0): First release on GitHub. Zenodo. https://doi.org/10.5281/zenodo.375923

Lœvenbruck, H. (s. d.). Praat scripts. Laboratoire de Psychologie et Neurocognition. https://lpnc.univ-grenoble-alpes.fr/spip.php?page=credits&lang=fr

Sadowsky, S. (2019). Recursos de Praat. http://sadowsky.cl/praat-es.html

Toscano, J. (s. d.). Praat script archives. https://sites.google.com/site/praatscripts/

Vicenik, C. (2009, agost). Praat script resources. The UCLA Phonetics Lab. http://phonetics.linguistics.ucla.edu/facilities/acoustic/praat.html

Winn, M. (2019). Praat scripts. http://mattwinn.com/praat.html

Manuals, guies i adjuda sobre la utilització de Praat

Eines per a l’anàlisi acústica de la parla: Praat

Praat Users List a Groups.io


Prosogram, P. Mertens, Department of Linguistics, KU Leuven


«Prosogram is a tool for the analysis and transcription of pitch variations in speech. Its stylization simulates the auditory perception of pitch by the listener. A key element in tonal perception is the segmentation of speech into syllable-sized elements, resulting from changes in the spectrum (sound timbre) and intensity. The tool also provides measurements of prosodic features for individual syllables (such as duration, pitch, pitch movement direction and size), as well as prosodic properties of longer stretches of speech (such as speech rate, proportion of silent pauses, pitch range, and pitch trajectory). The tool can easily interact with other software tools. It is used as the first step in automatic phonological transcription of intonation, the detection of sentence stress and intonation boundaries.

Processing steps:

The system is implemented as a Praat script.»


SegProso, J. M. Garrido, Departamento de Lengua Española y Lingüística General, Universidad Nacional de Educación a Distancia


«SegProso is a Praat-based tool for the automatic segmentation of speech corpora into prosodic units:

SegProso is currently available for Catalan, Spanish, Brazilian Portuguese and French. Other languages can be added with minimum or none adaptation of the scripts.

The tool uses a full rule and knowledge-based approach to perform the boundary detection tasks.

SegProso is available for public download.»


SFS, Speech Filing System, Speech, Hearing and Phonetic Sciences, University College London


«SFS 4/Windows is a free computing environment for PCs for conducting research into the nature of speech. It comprises software tools, file and data formats, subroutine libraries, graphics, special programming languages and tutorial documentation. It performs standard operations such as acquisition, replay, display and labelling, spectrographic and formant analysis and fundamental frequency estimation. It comes with a large body of ready made tools for signal processing, synthesis and recognition, as well as support for your own software development.

Analysis programs:

SFS is not public domain software, its intellectual property is owned by Mark Huckvale, University College London. However SFS may be used and copied without charge as long as the program remains unmodified and continues to carry this copyright notice.

Operating environments:
WIN32: Microsoft Visual C, WIN32 API. Windows 95/98/NT/2000.
Unix: GNU gcc compiler and X-Windows. SunOs, Solaris, Linux, etc.
MSDOS: Protected mode 32-bit with GNU compiler DJGPP. »


SFS/RTGRAM, Speech, Hearing and Phonetic Sciences, University College London


«RTGRAM is a free program for displaying a real-time scrolling spectrographic display of an audio signal. With RTGRAM you can monitor the spectro-temporal characteristics of sounds being played into the computer’s microphone or line input ports. RTGRAM is optimised for speech signals and has options for different sampling rates, analysis bandwidths, temporal resolution and colour maps.

RTGRAM is not public domain software, its intellectual property is owned by Mark Huckvale, University College London. However RTGRAM may be used and copied without charge as long as the program remains unmodified and continue to carry its copyright notice. »



«Signalyze is not currently available.

Signalyze is legacy voice and acoustic signal analysis software for the Macintosh developed by Eric Keller, Professor Emeritus, University of Lausanne, administered by Linguist Plus, and formerly distributed by Agora Language Marketplace. It requires Mac System 8, 9, or the Classic environment under OS X. It will not currently run without Mac OS 9 or the Classic environment..»


Sona, IKP, Institute for Communications Research and Phonetics, University of Bonn

«The program SONA is a versatile experimental tool for finding and visualizing relevant information in both the time and the frequency domain of a speech signal.

In the time domain, the program allows:

Furthermore, the segments can be marked and transcribed phonetically (Labeling).

In the frequency domain (lower half of the screen), the program generates a digital spectral analysis of the speech signal in 2D or 3D. The 3D representation of the time dependent power spectrum is known as Visible Speech or sonagram and is one of the most important practical tools of linguistics and phonetics. Sonagrams are represented in gray scale or colour coding in one of five frequency sections (0.5 to 8 KHz) with variable breadth. One mouse click enables the user to listen to a selected segment or measure frequency and intensity of its spectrum.»


Sonogram, C. Lauer

«Sonogram is a highly flexible audio spectrum analyzer for the analysis of sound, music and speech signals in the frequency-domain using different new algorithms.»


SoundIndex, M. Jacobson


«SoundIndex est un outil qui alie un éditeur de texte structuré en XML avec un éditeur de son. Il permet d’écrire des tags <audio> à n’importe quel niveau dans l’arborescence d’un fichier XML en mettant comme valeurs pour les attributs start et end celles qui sont lues dans l’éditeur de son. L’interprétation des tags <audio> se fait par le biais de feuilles de styles écrites en XSL.»


SoundScope, GW Instruments


«SoundScope software digitizes, analyzes, presents and databases speech and sound waveforms on Macintosh computers.

SoundScope is a third generation speech and sound analysis product line that represents a breakthrough in ease-of-use and advanced features.

Record a sound, perform analysis, extract key values, and compute statistics all with a few clicks of the mouse. Scroll through data, adjust the scale or display range, and even change the parameters for sound analysis computations.


Speech Analyzer, SIL International


«You can use Speech Analyzer to do the following tasks:

Operating system: Windows XP with Service Pack 3 (SP3), Windows Vista or Windows 7.»


SpeechStudio, Laryngograph Ltd.


«Speech Studio is a software and hardware package, which has been specially designed for phoneticians, speech scientists and quantitative work by ENT clinicians and SLT’s. It supports data recording direct to hard disk, real-time displays, and instantaneous quantitative analysis and pattern target mode for speech training.

Speech Studio software is Windows-based, user friendly, and feature rich.

Speech Studio also includes a very powerful program, which can make an extensive range of quantitative analysis on connected speech. It is seamlessly integrated with the data recording and display program. It can work on different kinds of speech pattern elements and produce powerful graph families. The speech elements include fundamental frequency, speech amplitude, contact quotient, nasality and friction.»


Transana, Spurgeon Woods LLC


«Transana is designed to facilitate the transcription and qualitative analysis of video and audio data. It provides a way to view video or play audio recordings, create a transcript, and link places in the transcript to frames in the video. It provides tools for identifying and organizing analytically interesting portions of video or audio files, as well as for attaching keywords to those video or audio clips. It also features database and file manipulation tools that facilitate the organization and storage of large collections of digitized video.»




«TranscriberAG is designed for assisting the manual annotation of speech signals. It provides a user-friendly graphical user interface (GUI) for segmenting long duration speech recordings, transcribing them, labeling speech turns, topic changes and acoustic conditions.

TranscriberAG is geared toward the needs of the speech research community, but its features might be found useful for other applications. It uses the Annotation Graph format as native format but can read a number of other annotation formats.

TranscriberAG is distributed as free software under the GNU General Public License GPLv3.»


TranscriberAG (Versió 2.0.0-b1) [Programa informàtic]. (2011). DGA. http://transag.sourceforge.net/


TranscriberAG (Versió 2.0.0-b1) [Programa informàtic]. (2011). DGA. http://transag.sourceforge.net/


WaveSurfer, Centre for Speech Technology, KTH


«WaveSurfer is an Open Source tool for sound visualization and manipulation. It has been designed to suit both novice and advanced users. WaveSurfer has a simple and logical user interface that provides functionality in an intuitive way and which can be adapted to different tasks. It can be used as a stand-alone tool for a wide range of tasks in speech research and education. Typical applications are speech/sound analysis and sound annotation/transcription. WaveSurfer can also serve as a platform for more advanced/specialized applications. This is accomplished either through extending the WaveSurfer application with new custom plug-ins or by embedding WaveSurfer visualization components in other applications.


WEDW Edit Waveform Program, Speech Research Lab, University of Delaware & A.I. duPont Hospital for Children


«Windows EDW (WEDW) is a fundamentally new program which attempts to provide similar functionality to the Unix/DOS version (EDW), but with a very different user interface.

WEDW retains some of the appearance of EDW in that a waveform display region is always present while spectrogram and pitch marking windows can be toggled on and off as desired. Both EDW and WEDW read and write waveforms in an extended RIFF (Microsoft .WAV) format that includes waveform segment definitions and both are also able to read an older .WAV format that was the original format used by EDW.

WEDW provides a way to display special symbols such as IPA phonetic symbols when a font for the symbols is available.

Prosodic features of duration, F0, and amplitude can be changed.»


WinCECIL, SIL International


«WinCECIL is a speech analysis tool based on the DOS CECIL version 2.1 program. WinCECIL provides support for recording, analyzing, and saving of 3 second sections of speech. WinCECIL requires a 20MHz 80386 computer or better running Microsoft Windows 3.1 or higher. It also requires a Windows Multimedia-compatible sound card.

Use this program to view speech recordings, automatic pitch contours, and spectrograms. Recording limit is 3 seconds.

Most of the functions of the WinCECIL program has been superseded by the Speech Analyzer program. This product has been discontinued and is no longer supported.»


Winpitch, Pitch Instruments


All these versions run under Windows® XP, Vista, Windows 7 and Windows 8. They also run on a Mac computer with an appropriate Windows emulator installed (such as BootCamp)»


WinSnoori, Y. Laprie, Université de Lorraine


«For several years we have undertaken the development of the software WinSnoori which is for both speech scientists as a research tool and teachers in phonetics as an illustration tool. It consists of five types of tools:


xassp, IPDS Institut für Phonetik und digitale Sprachverarbeitung, Christian-Albrechts-Universität, Kiel


«xassp is an application for displaying, analysing and processing speech signals. It is intended for segmental and prosodic labelling, but can be used for different purposes, because of its numerous configuration possibilities.

User-definable configurations allow to open several associated files together and to automatically perform certain analyses of the speech signal. The configuration Segmental, e.g. is intended for segmental labelling. The windows that are opened when choosing this configuration are:

The configuration Prosodic is used for prosodic labelling. When choosing this configuration the following windows are opened:

Although xassp is mainly intended for segmental and prosodic labelling, it provides several additional possibilities for analysing speech signals:



Eines per a l’anàlisi acústica de la parla

Eines per a la transcripció i l’etiquetatge de corpus orals

Eines per a l’anàlisi acústica i la transcripció de la parla
Joaquim Llisterri, Departament de Filologia Espanyola, Universitat Autònoma de Barcelona

La pàgina va ser modificada per darrera vegada el