Rationale: Treatment of oral and oropharyngeal cancer affects the anatomical regions involved in speech production. It has a major impact on patients' quality of life because it degrades communicational functions [1]. Despite this observation, little research is currently carried out on the functional impact of speech disorders. Previous studies have only found moderate correlations between quality of life and speech disorder severity scores, derived from a perceptual assessment [2]. This might be explained by the biases associated with standard perceptual speech evaluation, in terms of intra- and inter-rater reliability. In this context, the development of tools for automatic speech signal processing seems to be fully relevant [3]. Purpose: The objective of this study is to analyze the correlations between patients' quality of life and the speech disorder severity scores, assessed perceptually and automatically.Methods: The database used was the one from the C2SI project [2]. Eighty-seven patients treated for oral or oropharyngeal cancer were recruited. They filled out an overall quality-of-life questionnaire including a "social role functioning" dimension (MOS SF-36 "SF36-SF", [4]), but also two specific questionnaires related to speech [5, 6]. Their speech was then recorded on different tasks. A jury of six experts perceptually assessed the quality of speech and gave a score for the severity of the speech disorder on a semi-spontaneous task of picture description (ranging from 0 – no impairment – to 10 – maximum impairment). An automatic speech signal analysis allowed the calculation of a speech severity score (on pseudowords repetition, text reading and a sustained vowel "a"): the Carcinologic Speech Severity Index (C2SI). The correlations of the two severity scores (perceptual and automatic) with the scores from the quality of life questionnaires were studied using Spearman correlation coefficients.Results: The "social role functioning" dimension of the SF-36 (SF36-SF) was weakly correlated with perceptual (r =.13) and automatic (r =.31) scores. Regarding the speech-related quality of life, the functional dimensions of the SHI and of the PHI are moderately correlated with the perceptual score (r = -.39 for both questionnaires), but also with the automatic one (r = -.31 in both cases). The correlations between the speech severity scores and the functional dimensions of the questionnaires are also moderate (r = -.50 between the perceptual score and the PHI "psychosocial” dimension; r = -.52 between the automatic C2SI score and the "psychosocial” dimension).Discussion: This study shows that the correlation between the quality of life and the speech disorder severity is only moderate, whether the impairment is assessed perceptually or automatically. This requires studying the intermediate step between the speech disorder severity and the speech-related quality of life, which is the functional impact of the disorder on the patient’s everyday life activities. To this end, the development of comprehensive models combining different acoustic measures to avoid perceptual biases would allow for a better consideration of this functional impact. Further analyses have to be carried out to find whether these measures on conversational speech are relevant, which is a more natural context of speech production. The alteration of the prosody and the lexical choice according to the articulatory/phonetic complexity or to the syntactic structure could be cues of the subject’s speech behavior regarding avoidance or compensation strategies. Moreover, an automatic analysis of the speech signal including functional impact measurements could make it possible to determine thresholds for speech disorder severity according to the functional impact on communicational acts. In addition, the development of innovative objective tools that take into account the communicational impairment will improve current clinical practice by customizing patient care in order to meet individual daily functional needs starting from an early stage.