Separation of Speech from Simultaneous Talkers
نویسنده
چکیده
The separation of speech from two simultaneous talkers is a problem of some practical and theoretical importance. We describe a prototype separation system based on harmonic selection using comb filters. Hermes’ subharmonic spectrum method is used to produce a number of (weighted) pitch estimates, with pitch tracks for the two talkers then found by constrained dynamic programming. The system has successfully separated composite male/female /hVd/ tokens but performance is currently rather variable.
منابع مشابه
Three simultaneous speech recognition by integration of active audition and face recognition for humanoid
This paper addresses listening to three simultaneous talkers by a humanoid with two microphones. In such situations, sound separation and automatic speech recognition (ASR) of the separated speech are difficult, because the number of simultaneous talkers exceeds that of its microphones, the signal-to-noise ratio is quite low (around -3 dB) and noise is not stable due to interfering voices. Huma...
متن کاملOnline blind speech separation using multiple acoustic speaker tracking and time-frequency masking
Separating speech signals of multiple simultaneous talkers in a reverberant enclosure is known as the cocktail party problem. In real-time applications online solutions capable of separating the signals as they are observed are required in contrast to separating the signals offline after observation. Often a talker may move, which should also be considered by the separation system. This work pr...
متن کاملThe effects of single and double hearing protection on the localization and segregation of spatially-separated speech signals.
Recent results have shown that auditory localization in the horizontal plane is dramatically worse for listeners wearing double hearing protection ~earplugs and earmuffs! than it is for listeners wearing single hearing protection ~earplugs or earmuffs alone!. This suggests that double hearing protection might also impair the spatial unmasking that normally occurs when two simultaneous talkers a...
متن کاملImproving Multitalker Speech Communication with Advanced Audio Displays
Historically, most of the metrics that have been used to evaluate the effectiveness of military communications systems have focused on measuring the intelligibility of a single talker in the presence of a continuous noise masker. However, many critical military operations involve complex communications tasks that require listeners to monitor, process, and respond to two or more simultaneous spe...
متن کاملImprovement of three simultaneous speech recognition by using AV integration and scattering theory for humanoid
This paper presents improvement of recognition of three simultaneous speeches for a humanoid robot with a pair of microphones. In such situations, sound separation and automatic speech recognition (ASR) of the separated speech are difficult, because the number of simultaneous talkers exceeds that of its microphones, the signal-to-noise ratio is quite low (around -3 dB) and noise is not stable d...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2001